Social Networks Science Design Implementation

Security and Challenges Nilanjan Dey download

https://textbookfull.com/product/social-networks-science-design-
implementation-security-and-challenges-nilanjan-dey/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Intelligent Techniques in Signal Processing for

Multimedia Security 1st Edition Nilanjan Dey

https://textbookfull.com/product/intelligent-techniques-in-
signal-processing-for-multimedia-security-1st-edition-nilanjan-
dey/

U Healthcare Monitoring Systems Volume 1 Design and

Applications 1st Edition Nilanjan Dey

https://textbookfull.com/product/u-healthcare-monitoring-systems-
volume-1-design-and-applications-1st-edition-nilanjan-dey/

Acoustic Sensors for Biomedical Applications Nilanjan

Dey

https://textbookfull.com/product/acoustic-sensors-for-biomedical-
applications-nilanjan-dey/

Big Data Analytics for Intelligent Healthcare

Management 1st Edition Nilanjan Dey

https://textbookfull.com/product/big-data-analytics-for-
intelligent-healthcare-management-1st-edition-nilanjan-dey/
Applications of Firefly Algorithm and its Variants Case
Studies and New Developments Nilanjan Dey

https://textbookfull.com/product/applications-of-firefly-
algorithm-and-its-variants-case-studies-and-new-developments-
nilanjan-dey/

Security in Iot Social Networks Fadi Al-Turjman

https://textbookfull.com/product/security-in-iot-social-networks-
fadi-al-turjman/

Security surveillance centers : design, implementation,

and operation 1st Edition Disalvatore Anthony V

https://textbookfull.com/product/security-surveillance-centers-
design-implementation-and-operation-1st-edition-disalvatore-
anthony-v/

LTE communications and networks : femtocells and

antenna design challenges First Edition Safdar

https://textbookfull.com/product/lte-communications-and-networks-
femtocells-and-antenna-design-challenges-first-edition-safdar/

Industrial IoT Challenges Design Principles

Applications and Security Ismail Butun

https://textbookfull.com/product/industrial-iot-challenges-
design-principles-applications-and-security-ismail-butun/
Nilanjan Dey · Rosalina Babo
Amira S. Ashour · Vishal Bhatnagar
Med Salim Bouhlel Editors

Social Networks
Science: Design,
Implementation,
Security, and
Challenges
From Social Networks Analysis to Social
Networks Intelligence
Social Networks Science: Design, Implementation,
Security, and Challenges
Nilanjan Dey Rosalina Babo
•

Amira S. Ashour Vishal Bhatnagar

•

Med Salim Bouhlel

Editors

Social Networks Science:

Design, Implementation,
Security, and Challenges
From Social Networks Analysis to Social
Networks Intelligence

123
Editors
Nilanjan Dey Vishal Bhatnagar
Department of Information Technology Department of Computer Science
Techno India College of Technology and Engineering
Kolkata, West Bengal Ambedkar Institute of Advanced
India Communication Technologies
and Research
Rosalina Babo New Delhi
Department of Information Systems, India
ISCAP
Porto Polytechnic Med Salim Bouhlel
Porto Research Lab Sciences and Technologies
Portugal of Image and Telecommunication
Sfax University
Amira S. Ashour Sfax
Department of Electronics and Electrical Tunisia
Communications Engineering,
Faculty of Engineering
Tanta University
Tanta
Egypt

ISBN 978-3-319-90058-2 ISBN 978-3-319-90059-9 (eBook)

https://doi.org/10.1007/978-3-319-90059-9
Library of Congress Control Number: 2018942160

© Springer International Publishing AG, part of Springer Nature 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by the registered company Springer International Publishing AG
part of Springer Nature
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface

Social network analysis is a process of quantitative and qualitative analysis of vast

amount of data using different analytical techniques. It measures and maps the flow
of relationships and relationship changes between knowledge-possessing entities.
Online social networks are rapidly changing our lives. Their growing perva-
siveness and the trust that we develop in online identities provide us with a new
platform for security applications. Additionally, the integration of various sensors
and mobile devices on social networks has shortened the separation between one’s
physical and virtual (i.e., Web) presences. We envisage that social networks will
serve as the portal between the physical world and the digital world. However,
challenges arise when using social networks in security applications; for example,
how can one prove to a friend (or Friend) that your Facebook page belongs to you
and not a man in the middle? Once you have proved this, how can you use it to
create a secure channel between any device belonging to you and one belonging to
your friend?
The book is a collection of the eight chapters which are written by eminent
professors, researchers, and Industry people from different countries. The chapters
were initially peer-reviewed by the editorial board members, reviewers, and
industry people who themselves span over many countries. The whole book is
divided into three parts, namely Part I Social Network Design and Implementation:
An Overview, Part II Social Network Application: An Introduction, and Part III
Social Network Security and Challenges.
Chapter 1 by Md. Lizur Rahman, Md. Golam Sarowar, and Md. Sarwar
Kamal revolves around the sentiment classification. Nowadays social networks
generate a huge data from user’s view, emotions, thoughts, opinions, suggestions
regarding different products, events, places, brands, politics, etc. Those data play an
important role in different ways. Technically, in the interval of every 60 s in a social
network like Facebook, lots of comments and statuses are updated which are
associated with thousands of contexts. However, realization of different ways in
which texts are seems to be appeared on Facebook can help us to improve our
products. In general, different organizations such as text organization used senti-
mental analysis for successful classification. They transpired feelings, emotions in

v
vi Preface

different forms like positive, negative, friendly, and unfriendly. To solve this
problem, authors have concentrated on different techniques of deep learning. In this
paper, authors highlighted few deep learning implementation techniques known as
convolutional neural network and recursive neural network with classification of
different texts.
Chapter 2 by Nabanita Das, Surekha Borra, Nilanjan Dey, and Samarjeet
Borah enlightened that movie recommendation systems are a common practice by
most of the online stores today. The Web-based movie recommendation systems
make predictions about the responses of the users based on their search history or
known preferences. Recommendation of items is usually done based on the prop-
erties or content of the item or collaboration of the user’s ratings, and by using
intelligent algorithms that include classification or clustering techniques. Accurate
prediction of what the customer may likely to busy or the user my visit is of utmost
important, as it benefits both the service providers and customers. This chapter
provides the evolution, fundamental concepts, classification, traditional and novel
models, requirements, similarity measures, evaluation approaches, issues, chal-
lenges, impacts due to social networking, and future of movie recommendation
systems.
Chapter 3 by Sapna Sinha, Vishal Bhatnagar, and Abhay Bansal analyzed
that data available on Twitter can be in form of text, photographs, customer pref-
erences can be identified using Twitter analytics which can help service providers to
offer personalized services. If tour operator is able to predict trends, they can easily
set optimized price and prepare well in advance to provide unforgettable trip to their
customers. Tour operators adopt list pricing policy for deciding price of the tourism
product and also there is no set model available for this. The tour operators set the
price which helps them to gain high profit, but due to non-availability of any
standard formula the decided price varies with the price offered by competitors.
Prices are kept high when season is at the peak and more and more tourists are
visiting the place or purchasing the tourist products, similarly price is kept low
when season is low. In this chapter, authors had proposed pricing model consid-
ering different factors that decides rates of the product in the tourism sector.
Real-time analytics performed on the data available on the Web portals or social
networking sites are used to get the most trending tourist destination, and the tour
operators functioning at different destination can set price of their products using
the proposed model. Real-time analytics will help tour operators to analyze the
demand in coming season.
Chapter 4 by Arushi Jain and Vishal Bhatnagar discussed that social media is
rapidly providing new standards of interaction between individuals in the recent era.
It is known as a computer-intermediated tool that tolerates people to share, or create
information, medical notes/reports, ideas, and pictures/videos through virtual
communities. The online citizens (netizens) are able to create a colossal network of
people to communicate with. Social media such as Facebook, Twitter, YouTube,
Instagram, and Tinder have prevalent uses that produce copious amount of data
which is beyond the ability of normal software tools to process in the given elapsed
time. Apache Hadoop project is the most famous open-sourced frameworks for
Preface vii

large-scale computation on the commodity hardware. Hadoop has become kernel

for distributed operating system for big data. There are two core components
associated with Hadoop—Hadoop Distributed File System (HDFS) and
MapReduce. MapReduce distributed the tasks on multiple nodes in the cluster; the
developer only have to write code, rest is taken care by MapReduce. The generated
data from these social sources is real time and includes information about author’s
daily activities, feelings, and emotions. The messages often include images,
geolocations, and many other annotations. This vast data repository provides
researchers with opportunities to study the individuals’ behavior/emotions that
subject to different conditions. In this chapter, authors found the trending tweets
from January 2017 to September 2017 depending upon the eight prominent themes
that emerged from the dataset and trending tweets depending upon the geolocation.
For this, author divided India depending upon the region, that is North India, West
India, South India, East India, Central India, and Northeast India.
Chapter 5 by C. Sardianos, N. Tsirakis, and I. Varlamis discussed that it is
typical among online social networks’ users to share their status, activity, and other
information with fellow users, to interact on the information shared by others, and
to express their trust or interest for each other. The result is a rich information
repository which can be used to improve the user experience and increase their
engagement if handled properly. In order to create a personalized user experience in
social networks, we need data management solutions that scale well on the huge
amounts of information generated on a daily basis. The social information of an
online social network can be useful both for improving content personalization and
for allowing existing algorithms to scale to huge datasets. All current real-world
large-scale recommender systems have invested on scalable distributed database
systems for data storage and parallel and distributed algorithms for finding
recommendations. This chapter focuses on collaborative filtering algorithms for
recommender systems and briefly explains how they work and what their
limitations are.
Chapter 6 by A. L. Chapman, M. Lei, and C. Greenhow analyzed the design,
implementation, benefits, and challenges of using a Facebook application, College
Connect, are presented. College Connect was designed to address the persistent
educational problem of college access in the USA, part of which stems from stu-
dents’ lack of social capital, the human and information resources available to them
in their social networks that can provide needed information, such as how to apply
to, enroll in, and pay for college. College Connect, a social networking application
which runs on Facebook, the parent platform, was designed to help address this
problem by creating a network visualization of each student’s Facebook Friends
network and showing the student who within the network has college information
in their Facebook profile. In this chapter, authors explained the theory and proce-
dures that led to the design of the College Connect application, the process of
launching the application, and the benefits and challenges of implementing it with
adolescent students preparing for college.
viii Preface

J. G. Tromp, Chung Le Van, Bao Le Nguyen, and Dac-Nhuong Le in

Chap. 7 covered the convergence of social media networks and virtual reality
systems, labeled as social virtual reality. It reviews the evolution of the World Wide
Web from a single-user, static experience into the futuristic 3D multiuser interactive
experience. This is followed by a review of bulk data collection in virtual reality,
and the ethical risks and threats to privacy that this could create for social virtual
reality users. The chapter ends with recommendations to mitigate the ethical risks
and threats to privacy, for adult VR users, parents of VR users, psychologists, VR
software, and hardware manufacturers.
Chapter 8 by I. Kožuh and M. Debevc provided a comprehensive insight into
the use of social media among the deaf and hard of hearing, along with the benefits
and challenges in use. Existing recommendations toward overcoming the chal-
lenges are reviewed, and approaches for design of social media and its efficient use
are proposed. The findings may serve social media developers, educators, social
inclusion advisors, and policy makers on how to apply social media as an inclusive
tool for participation in society.

Kolkata, India Nilanjan Dey

Porto, Portugal Rosalina Babo
Tanta, Egypt Amira S. Ashour
New Delhi, India Vishal Bhatnagar
Sfax, Tunisia Med Salim Bouhlel
Contents

Part I Social Network Design and Implementation: An Overview

1 Teenagers Sentiment Analysis from Social Network Data . . . . . . . . . 3
Lizur Rahman, Golam Sarowar and Sarwar Kamal
2 Social Networking in Web Based Movie
Recommendation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Nabanita Das, Surekha Borra, Nilanjan Dey and Samarjeet Borah
3 Sifting Through Hashtags on Twitter for Enterprising Tourism
and Hospitality Using Big Data Environment . . . . . . . . . . . . . . . . . . 47
Sapna Sinha, Vishal Bhatnagar and Abhay Bansal
4 Hashtag# Perspicacity of India Region Using Scalable Big Data
Infrastructure Using Hadoop Environment . . . . . . . . . . . . . . . . . . . . 67
Arushi Jain and Vishal Bhatnagar

Part II Social Network Application: An Introduction

5 A Survey on the Scalability of Recommender Systems
for Social Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
C. Sardianos, N. Tsirakis and I. Varlamis
6 Social Network Applications for Education: The Case
of College Connect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
A. L. Chapman, M. Lei and C. Greenhow

ix
x Contents

Part III Social Network Security and Challenges

7 Massively Multi-user Online Social Virtual Reality Systems:
Ethical Issues and Risks for Long-Term Use . . . . . . . . . . . . . . . . . . 131
Jolanda Tromp, Chung Le, Bao Le and Dac-Nhuong Le
8 Challenges in Social Media Use Among Deaf and Hard of Hearing
People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
I. Kožuh and M. Debevc
About the Editors

Nilanjan Dey is an Assistant Professor at the Department of Information

Technology, Techno India College of Technology, Kolkata, WB, India. He holds
an honorary position of Visiting Scientist at Global Biomedical Technologies Inc.,
CA, USA, Research Scientist of Laboratory of Applied Mathematical Modeling in
Human Physiology, Territorial Organization of Scientific and Engineering Unions,
Bulgaria, and Associate Researcher of Laboratoire RIADI, University of Manouba,
Tunisia. His research topics are medical imaging, soft computing, data mining,
machine learning, rough set, computer-aided diagnosis, and atherosclerosis. He has
20 books and 300 international conferences and journal papers. He is the editor in
chief of International Journal of Ambient Computing and Intelligence (IGI Global),
USA; International Journal of Rough Sets and Data Analysis (IGI Global), USA;
the International Journal of Synthetic Emotions (IJSE), IGI Global, USA; and
International Journal of Natural Computing Research (IGI Global), USA, Series
Editor of Advances in Geospatial Technologies (AGT) Book Series, (IGI Global),
USA; Executive Editor of International Journal of Image Mining (IJIM),
Inderscience, and Associated Editor of IEEE Access Journal and the International
Journal of Service Science, Management, Engineering and Technology, IGI Global.
He is a life member of IE, UACEE, ISOC.

Rosalina Babo is a Coordinator Professor at the School of Accounting and

Administration of Porto/Polytechnic of Porto (ISCAP/IPP), Portugal. Since the year
2000, she is the head of the Information Systems Department and was a member
of the university scientific board for 12 years (2000–2012). Rosalina Babo was one
of the founders (2006) of CEISE/STI research center and its director until the year
2011. Having several published papers in international conferences and books, her
main areas of research are e-learning, usability, e-commerce, and social networks.

Amira S. Ashour is currently an Assistant Professor and Head of Department-

EEC, Faculty of Engineering, Tanta University, Egypt. She has been the Vice Chair
of Computer Engineering Department, Computers and Information Technology
College, Taif University, KSA, for one year from 2015. She has been the Vice

xi
xii About the Editors

Chair of CS department, CIT college, Taif University, KSA, for five years. Her
research interests are smart antenna, direction of arrival estimation, targets tracking,
image processing, medical imaging, machine learning, signal/image/video pro-
cessing, image analysis, computer vision, and optimization. She has 6 books and
about 70 published journal papers. She is an editor in chief of the International
Journal of Synthetic Emotions (IJSE), IGI Global, USA. She is an Associate Editor
for the IJRSDA, IGI Global, USA, as well as the IJACI, IGI Global, USA. She is an
editorial board member of the International Journal of Image Mining (IJIM),
Inderscience.

Vishal Bhatnagar holds B.Tech., M.Tech., and Ph.D. in the engineering field. He
has more than 18 years of teaching experience in various technical institutions. He
is currently working as a Professor in Computer Science in Engineering
Department, Ambedkar Institute of Advanced Communication Technologies and
Research (Government of Delhi), GGSIPU, Delhi, India. His research interests
include database, advance database, data warehouse, data mining, social network
analysis, and big data analytics. He has to his credit more than 110 research papers
in various international/national journals and conferences. He is also member of
many international journals as editorial board, editor or reviewer board member. He
has been guiding the postgraduate and Ph.D. students in the field of data science
from 2010.

Med Salim Bouhlel was born in Sfax (Tunisia) in December 1955. He is a full
professor at Sfax University, Tunisia. He is the Head of the Research Group:
Sciences and Technologies of Image and Communication since 2003.
He was the Director of the Higher Institute of Electronics and Communications
of Sfax—TUNISIA (ISECS) 2008–2011. He received the golden medal with the
special appreciation of the jury in 1999 on the occasion of the first International
Meeting of Invention, Innovation and Technology (Dubai, UAE). He was the Vice
President and Founder Member of the Tunisian Association of the Specialists in
Electronics and the Tunisian Association of the Experts in Imagery. He is the
President and Founder of the Tunisian Association on Human–Machine Interaction
since 2013. He is the editor in chief of the International Journal “Human-Machine
Interaction” and a dozen of special issues of international journals. He is the
chairman of many international conferences and member of the program committee
of numerous international conferences. His research interests are image processing,
telecommunication, and human–machine interaction in which he has obtained more
than 20 patents so far. More than 400 articles were published in international
journals, conferences, and books. Moreover, he has been the principal investigator
and the project manager for several research projects dealing with several topics
concerned with his research interests mentioned above.
 Part I
Social Network Design and
Implementation: An Overview
Chapter 1
Teenagers Sentiment Analysis from
Social Network Data

Lizur Rahman, Golam Sarowar and Sarwar Kamal

Abstract Now a day’s social networks generate a huge data from user view, emo-
tions, thoughts, opinions, suggestions regarding different products, events, places,
brands, politics etc. Those data plays an important role in different ways. Techni-
cally, in the interval of every 60 s in a social network like Facebook, lots of comments
and statuses are updated which are associated with thousands of contexts. However,
realization of different ways in which texts are seems to be appeared on Facebook
can help us to improve our products. In general, different organizations such as text
organization used sentimental analysis for successful classification. They transpired
feelings, emotions in different form like positive, negative, friendly, unfriendly etc.
To solve this problem we have concentrated on different techniques of deep learning.
In this paper we highlight about few deep learning implementation techniques known
as Convolutional Neural Network and Recursive Neural Network with classification
of different texts.

Keywords Facebook · Sentimental analysis · Deep learning · Convolutional

neural network · Convolutional hidden layer · Max pooling · Softmax layer
Recurrent neural tensor network

1.1 Introduction

Sentiment classification is one of the most important tasks in social network analysis
in current circumstance sentiment analysis for various social networking sites seems

L. Rahman · G. Sarowar · S. Kamal (B)

Department of Computer Science and Engineering, East West University,
Dhaka, Bangladesh
e-mail: [email protected]
L. Rahman
e-mail: [email protected]
G. Sarowar
e-mail: [email protected]

© Springer International Publishing AG, part of Springer Nature 2018 3

N. Dey et al. (eds.), Social Networks Science: Design, Implementation,
Security, and Challenges, https://doi.org/10.1007/978-3-319-90059-9_1
4 L. Rahman et al.

to be a very potential research equipment. The main reason behind this includes
online user’s free authorization and assessment to their account whenever they want.
Various recent surveys demonstrate that companies collect huge amount of data
through the web to extend their customers satisfaction analysis. Setters can also
watch terrorist’s activity through collecting data from web to extend sentimental
analysis and are getting aware and taking steps about their baleful events. The authors
used sentimental analysis to find the imaginative sentiment automatically from uses
reviews [1]. Sentiment analysis is a natural language processing problem where
text is understood and the underlying intent is predicted. Few authors have already
published lots of contributions associated with sentimental analysis. Most of the
contributions have been accomplished targeting to twitter, product reviews, political
party, tweet updates, brands of products [2]. These system expresses a positive or
a negative opinion(e.g., blog post, product review, etc.) about a given entity(e.g.,
product, policy, person etc.). Before imposing appropriate mechanism over data they
collect raw data from twitter or facebook and apply their implemented methods on
the data they gathered. Since, twitter has a restriction of using 140 characters of
total length while posting something in twitter [2], thus users supposed to get narrow
scope to express their thoughts, expressions, emotions.
In general, recent years highly keep in touch with available social networks and
collaborative technologies. Thousands of research associated with social media has
been conducted already. However, various study demonstrate that this social media
and collaborative technology can negotiate with traditional way of learning into
a collaborative one [3]. Meanwhile, social media seems to be effective in health
care technology areas also. For the time being health care is becoming sharing and
cooperative which makes realizing about the necessity of social media technologies.
Moreover, for present time online based social networking sites are providing the
platform for the people to express their opinions, thoughts, emotions, short messages
etc. within a second of click. With the help of social networks nowadays it’s easier to
be interconnected. This phenomenon results with effective outcome. However, with
the increasing effectiveness of social networks and collaborative technologies the
data passing through various platforms associated with these appeared to be a great
evolution. Enormous posts, comments, emotions, reactions etc. are need to be manip-
ulated with the shortest possible time. Despite all of this present circumstances of
various social networks negotiate with lots of real life conditions among which adver-
tisement of various products and prospects are comparatively most significant. Thus,
social networks play an important role in this era providing multimedia enriched
data, aggregated feedback and personal information sharing. Among all the social
networks Facebook is considered most popular one by connecting every stages of
people together within a single shade. According to Wikipedia Currently Facebook
is dealing with approximately 2 billion people, twitter with 319 million users, 375
million users. However, the researchers are mainly concerns with determining exact
sentiment, emotions, and reactions filtering lots of multimedia based data. Thus, this
study is concerned with this issue.
In this paper, in most cases we have concentrated on Facebook as our targeted
website. From the perspective of virtual world status is creating an updated feature
1 Teenagers Sentiment Analysis from Social Network Data 5

Fig. 1.1 Graphical

representation of tweets rate
of growth [55]

of regular activity, thoughts, emotions etc. From users side. Facebook has a total
number of 5000 characters for updating every status which has a big difference from
Twitter [4]. Therefore, the main advantage is people transpired emotions clearly
which is desirable. Total Facebook users are enormous and so it is much easier to
collect samples. Analyzing these samples we can get identify user’s motion of interest
which would be good enough. Realization of different ways in which texts are seems
to be appeared on Facebook can help us to improve our products.
According to June 2017, a survey [5] on Facebook statistics shows Every 60 s on
Facebook: 510,000 comments are posted, 293,000 statuses are updated. Understand-
ing the various ways texts are used on Facebook can help us to improve our products.
In general, whenever a user expresses his views, it becomes important for the com-
munity to correctly identify the requirements of user to make him stay longer. By the
analysis of product reviews given by the customers, an organization can decide about
the future circumstance on any particular product. Thus, it becomes mandatory for
the organizer to analyze the comments given in social media to extract any essential
information from huge volume of data. However, the annual tweet growth rate for
twitter from 2006 to 2013 [6] is followed by Fig. 1.1.
In general, sentiment analysis or opinion mining is a most hotly debated topic
nowadays. This is basically the computational illustration of people’s comments,
attributes, comments, posts, and tweets on any specific issue. Those issues can
demonstrate both individuals as well as events and topics. However, both the topic
sentiment analysis and opinion mining refers to interchangeable circumstances of
some issues. Despite of this some of the researchers in this field who have con-
ducted various contributions have estimated that both of the term SA (sentiment
analysis) and OM (opinion mining) indicates different motion rather than indication
same dimension. According to their derivation, sentiment analysis is such kind of
machine learning approach which works with opinion mining whereas sentiment
analysis refers to the sentiment explicated on various issues for this context in var-
6 L. Rahman et al.

ious tweets. Therefore, according to sentiment analysis this specific topic illustrate
opinions regarding a topic or event then identifying the sentiment expressed by the
users and determining their polarity. Basically, sentiment analysis overall process is
the combination of three kinds defined as document-level, sentence-level, and aspect-
level [7]. Initially, the given sentence will be evaluated whether it is a subjective or
objective sentence. Thereafter, SA will determine which sentiment is depicted by
the sentence given. Since current world is stepping towards the evolution of higher
dimensional, complex and complicated data, it will be main obstacle for the previous
methods to cope with new dimension of data. Moreover, various social network asso-
ciated data like those information produced from twitter, Facebook, Google etc. are
increasing in an exponential manner and for which it demands strong fundamental
machine learning approaches to be in action to manipulate with better accuracy and
efficiency [8]. Realizing the worth of this issue we have concentrated on convolutional
neural network as well as Recursive tensor neural network for sentiment analysis and
text mining. Our derived approaches represents better performance for conducting
this task. However, for this study we only highlight on English data classification or
analysis. Our future contribution may include all the available language data analy-
sis. However, collected data can be noise full, irreverent which are considered to be
the main obstacle for the context of accurate analysis and efficient classification of
sentiment provided by the users un some specific events, topics etc. therefore, various
mapping algorithms like self-organization map, Ordinary Least Squares Regression
(OLSR), Linear Regression., Logistic Regression., Stepwise Regression., Multivari-
ate Adaptive Regression Splines (MARS), Locally Estimated Scatter plot Smoothing
(LOESS) those algorithms are highly appreciated. Thus, our future contribution will
include mapping based machine implementation of mapping based machine learning
algorithms for sentiment detection and opinion mining.

1.2 Literature Review

Many authors applied and contributed different methods for performing sentimental
analysis. Alongside, lots of researchers have used deep learning, machine learning,
vector machine etc. In literature, lots of outperformers’ methods have been exposed
through various researchers in paragraph and sentence level. In [9], the contributors
have drawn their attention towards minimum cut in text graphs to achieve the main
partition of the words from the whole text. Moreover, though the use of machine
learning methods they have performed sentiment analysis on the selected texts only.
In [10], the authors have embedded convolutional neural networks to train handwrit-
ten digits. The researchers of this work have been able to take their performance to
the highest level though demonstration of 99.6% accuracy. In [11] authors have used
different techniques to reduce over fitting and they have improved the performance to
a top level up to 99.67%. They have emphasized on 10,000 MNIST test images (seen
except training session) and the system defined by them determine 9,967 correctly.
From rest 33 images many of those are tough even for a human being to differ-
1 Teenagers Sentiment Analysis from Social Network Data 7

entiate. They used an ensemble of networks and build a model to improve entire
performance. In [12], the contributors mentioned support vector machine mecha-
nism for better classification from various data source collected from specific social
networks. In [3], initially the authors have classified texts into neutral and polar first
then have determined whether the text is expressed as positive or negative senti-
ment. In [13], the authors have depicted distant learning approach to differentiate
sentiment analysis data. They have emphasized on twitter’s every tweet’s conclud-
ing emotions. If concluding emotions are in positive emoticons like “:)”, “:-)” are
considered as positive and if rest emotions are appeared like “:(”, “:-(” then that will
be considered as negative. They build models using different classifiers and report
that Support Vector Machines (SVM) outperforms other classifiers comparatively.
They also have employed different models in terms of feature space and they report
that Unigram model outperforms all other models. Moreover, according to the con-
tribution mentioned in [14] overall sentiment has been taken in account rather than
concentrating on topic based analysis which is fruitful when overall input datasets
are noise free or irreverent free. But from the perspective of big data analysis or
noise full data this contribution can be detected as slower prediction mechanism. In
this respect of view our contributed approach works better even after the presence of
noise or irreverent data. Another contribution regarding sentiment analysis [15] indi-
cate the use of corpus for sentiment analysis as well as decision or opinion mining.
This contributions reveals that their proposed approach represents higher accuracy
along with efficiency for twitter data analysis. Initially for this they have drawn their
attention towards prediction of positive, negative and neutral sentiments for every
document they have in English language. The authors also demonstrate that their
future work will be conducted on all the language available currently. Since our con-
tribution highlights on various neural network based algorithm to train the datasets,
thus overall accuracy as well as efficiency outperforms its alternatives. Additionally,
the work [16] has explored lots of work associated with sentiment analysis in recent
time. Details discussion of various works create a new way to know gist content in
a while.
More research on sentiment analysis along with big data has been also conducted
by thousands of researchers. Among them the contribution [17] deals with vast
amount of sentiment rich data or texts through status updates, reactions, comments,
likes, tweets etc. the authors of this paper have highlighted into analysis of sentiment
for electronic devices including mobiles, Tabs, laptops etc. to be user friendly than
before. They have demonstrated that through the use of sentiment analysis in specific
domain the effect of domain information can be extracted easily. Along with analyz-
ing vast amount of data the authors have also implemented a new feature defined as
vector classifier for determination of more accuracy. According to the contribution of
[18], the authors in this study mostly concentrated on feature selection, negation deal-
ing and emotion handling. For the purpose of increasing performance of machine
learning algorithms this study negotiates with huge numbers of text datasets like
tweets or movie reviews and evaluates the performance of machine learning algo-
rithms. Thus, evaluation of public opinion as well as sentiment analysis has been taken
into account in this study for enhancement of available machine learning algorithms
8 L. Rahman et al.

in this field. Intuitively, the study [19] refers to illustration or details demonstration
of those datasets which are increasing in an exponential manner including contents
on the Web like twitter, amazon, Facebook etc. since these types of large volume of
data can create a great harm for the efficiency as well as accuracy and sensitivity of
machine leaning algorithms. From the perspective of contribution [20] lots of papers
regarding sentiment analysis have been explored in details for probable acquisition
of whole knowledge at a time from one paper. Moreover, nowadays Facebook, twit-
ter, amazon and other social websites support the feature of using various slang, Gif
images, emotions, reactions, comments which are mainly responsible for produc-
tion of higher dimensional large volume of datasets. This study also concentrates
on this issue and demonstrates the machine learning ways through which we can
resolve these types of problems. Since nowadays the datasets are increasing in an
exponential manner for which the classification task is becoming slower data by day.
Avoiding this situation various mapping based machine learning approaches can be
most helpful for classification, feature detection as well as to resolve inefficient sen-
timent analysis problems. In this respect [21–27] these study have proposed various
mapping less as well as mapping based algorithms which are capable of classifying
the data in a complex as well as complicated situation. However, another studies
mentioned in [28–37] highlights various automated machine learning system for
manipulation of imbalance, large volume of data and also to extract gist information
from data gathered from Facebook, tweeter data. Details performance evaluation
also demonstrates outperformances of those algorithms implemented and proposed
in those contributions. Moreover, [38–48] contributions highlight developing various
artificial intelligence algorithms which are capable of conducting various classifica-
tion as well as alignment task with higher accuracy and efficiency compared to other
alternatives in literature.

1.3 Deep Learning Algorithms (CNN) Used in Sentiment

Analysis

Basically, deep learning refers to combination of various algorithms. Such as Convo-

lutional Neural Network (CNN) extended version of neural network (NN), Recursive
Neural Tensor Network (RNTN) extended version of Recursive NN basically con-
cerned with the use of specific sparse tree for every input data, Recurrent Neural
Network (RNN) etc. those are very popular. Basically NN are very efficient for text
generation, image classification, pattern classification, vector representation, classi-
fication of sentence etc. [49]. Deep learning networks can be used for its automatic
learning capability and we can training for supervised as well as unsupervised both
[50]. Due to supervised and unsupervised learning techniques deep learning is very
useful. Many researchers used deep learning for handle sentiment analysis. Authors
used data from rottentomatoes.com and make the representation of movie review by
using Recursive Tensor Neural Network (RNTN) [51].
1 Teenagers Sentiment Analysis from Social Network Data 9

Fig. 1.2 Overall infrastructure of convolutional neural network

1.3.1 CNN (Convolutional Neural Network)

In general, convolutional neural network is the extended version of previously

revealed neural network. This new version of neural network includes one of the
amazing features defined as multiple perceptron. From the perspective of artificial
neural network this term multiple perceptron focuses on lesser preprocessing which
was not included in the previous neural network. Therefore, for this extraordinary
feature of convolutional neural network this mechanism performs better compared to
all the previous neural nets. Moreover, this specific capability which convolutional
neural nets owns inside itself makes the way easier by processing the input data
initially just to avoid redundant or irreverent data. Basically, this evolutionary algo-
rithm got concentrated on after successful interpretation of a well-known research
work explored by LeCun and Bottou [52]. This work illustrates CNN as inspired
by biological knowledge as well as mathematical theorem along with co-operation
of computer science. Therefore, for this study the convolutional neural network is
being focused for its extraordinary ability of dimensionality reduction along with
pre-processing strategy built in inside it. The basic infrastructure of convolutional
neural network is illustrated in Fig. 1.2.
Mathematical derivation of convolutional neural network illustrates a straight-
forward way of representing an overall process of machine learning algorithm. In
usual cases, the input matrix size of convolutional neural network covers the input
image or datasets successfully. However, the back-propagation paradigm is one of the
glamorous thing to be attracted by convolutional neural network. The general back
propagation algorithm for convolutional neural network is demonstrated as follows:
10 L. Rahman et al.

Thus, discussion conducted above represents the successful illustration of convo-

lutional neural network algorithm for the purpose of sentiment analysis. Here initial
phage depicts input datasets in a compacted matrix. Subsequently, computation of
i
px,y for each and every iteration becomes mandatory on the initial input datasets com-
pacted within a matrix. Meanwhile, estimated output error need to be concentrated
on for successfully implementation of back-propagation algorithm. If error seems to
be appeared then using back-propagation mechanism we need to propagate the error
and need to be minimized the error as much as possible. Now, gradient function is
usually used for achieving desired output from the overall CNN approach.
Basically, the input matrix of convolutional neural network covers overall input
data. Besides, internal calculation is simpler to define and easy to implement. For
interpretation of CNN if we have N x N input matrix and if we consider a K x K filter
ω then overall
 outcome
 matrix will be (N − K +1) × (N − K + 1). Thus, outcome
of 1st layer Oi j same time input of 2nd layer Oil j can be defined as follows,
l


K −1 K
 −1
Oil j  ωab y(i+a)(
l−1
j+b) (1.6)
a0 b0

However, outcome of convolutional neural layer is depicted using following equation

in practical,

Cil j  σ (Oil j ) (1.7)

 
∂E
Meanwhile, error calculation rate ∂C l which is defined as partial derivative of
ij
error rate (E) with respect to the outcome of previous calculation layer is mandatory
to summarize to determine all the output of each neuron lies on any layer. Therefore,
the equation is,
1 Teenagers Sentiment Analysis from Social Network Data 11

N−K N
−K N−K N
−K
∂ E ∂ Oi j
l
∂E ∂ E l−1
  C (1.8)
∂ωab i0 j0
∂ O l ∂ω
ij ab i0 j0
∂ Oil j (i+a)( j+b)

For further proceedings what we need to do is to compute the gradient. Computation

of gradient can be demonstrated as,
∂E ∂E ! l
 σ (Oi j ) (1.9)
∂ Oil j ∂Cil j

Above all, for achieving final result following mathematical chain rule can be
adjusted,

 −1 K
 −1  −1 K
 −1
∂E
K
∂E ∂ O(i−a)(
l
j−b)
K
∂E
  ωab (1.10)
∂Cil−1
j a0 b0
∂ O(i−a)(
l
j−b) ∂Cil−a
j a0 b0
∂ O(i−a)(
l
j−b)

Thus, above discussion bears the realization that Convolutional neural network
(CNN) follows a specific infrastructure which particularly has been marked as better
classifier. This mechanism is quite faster compared to other alternatives along with
faster training facility and also allows us to train deeply by forming lots of network
layers internally. Here for this context, we have explored some powerful techniques
including convolutions, pooling, dropout technique (to reduce unnecessary data) etc.
we have also used some integrating ideas such as softmax, back propagation, regu-
larization etc. CNN is a combinational feature of three basic ideas- local receptive
fields, shared weights and biases and pooling.
In usual cases, whenever a data is inserted into CNN it will be sent into differ-
ent layers (first layer is considered input layer, last layer is considered output layer
and in between first and last is hidden layers) subsequently and finally will produce
sentimental results (e.g., positive, negative, neutral, etc.). However, every layer con-
tains lots of neurons. For clarification we have considered that input layer contains
28*28 square of neuron and whose value corresponds to the 28*28 pixel. Basically,
each and every neuron of any layer is fully connected to the neurons of their neighbor
layer. However, in every local receptive field all the hidden neurons are not connected
under manual supervision, rather each neuron will be connected to a small region
of the input neurons. Graphical representation of local receptive field is as follows
Fig. 1.3,
Same matrix is taken into account while considering bias, then the same weights
and bias for each of the 24*24 hidden neurons will be applied to execute overall
process. For the j, k hidden neurons, output is-
4 4
C(b + wl,m a j+l, k+m ); δ  neural activation f unction,
i0 m0
b  shar ed value f or bias, wl,m  5 ∗ 5 array o f shar ed weights (1.11)
Very often Eq. 1.11 is defined as convolution. Shared weights and biases greatly
reduce total parameters involving in the convolutional network.
12 L. Rahman et al.

Fig. 1.3 Basic infrastructure of Local receptive field

Fig. 1.4 Generalized form after applying max pooling

Pooling mechanism is usually used immediately after convolutional layers. The

pooling mechanism usually filters the immediate output from the convolutional neu-
ral layer and also preprocess the resultant data. Each unit of a pooling layer mainly
summarize approximately 2*2(say) neurons from whatever we get from the previ-
ous layer. A common procedure for pooling can be characterized as max pooling
(Fig. 1.4).
After pooling Softmax layer will be in action automatically to remove the problem
of learning slowdown. This is basically used for faster as well as efficient learning
process. We use softmax to create a new type of connection in the final layer. Because
the connection between a hidden layer to next hidden layer is not fully connected. If
we do not change the connection as fully connected, output of any neuron could be
missed from last hidden layer.


L
j ez j
a Lj   L 1 (1.12)
j k ezk

where, a Lj  output activation for j neuron of L layer.

And z Lj  corresponding weighted input for j neuron of L layer.
This equation shows us output of activations for j’th neuron of L’th layer are
positive, because of positive exponential function and output range in between 0
1 Teenagers Sentiment Analysis from Social Network Data 13

Fig. 1.5 Overall basic infrastructure of this contribution

to 1(positive, negative, neutral). Thus, all the functions are combined together and
accomplished a convolutional neural network (CNN). This can be performed well
in both supervised and unsupervised learning. However, for sentiment analysis a
system has been proposed for twitter [53]. For this purpose they used convolutional
neural network with initialize the weight. While they avoiding requirement for new
features they find train accurately is critical (Fig. 1.5).
We will input a text into CNN and it will show the sentiment rate of positive,
negative and neutral by using sentimental algorithms, build up by CNN (Table 1.1).

1.3.2 Recursive Neural Tensor Network (RNTN)

In general, Recursive neural tensor networks (RNTNs) are some kinds of neural
nets which are exceptionally useful for all kinds of language processing including
natural language. They have specific tree structures corresponding to a neural net at
each node. Basically, Recursive neural networks can be significant for the purpose of
boundary segmentation, differentiation between positive and negative word groups
etc. for sequential classification word vectors are usually referred as features and
served as basic infrastructure, after which all the findings are compacted within
various sub groups and phases for final sentiment determination purpose. For this
contribution Recursive Neural tensor network performs better because of branching
factor it uses. However, RNTN (Recursive tensor Network) usually deals with some
specific structures or sparse tree for hierarchical processing of overall steps structured
in a tree. This specific algorithm seems to be working better while time is not being
concentrated on. This is basically a recursive chain process which follows one process
after another. Sentiment analysis is also possible through the use of convolutional
14 L. Rahman et al.

Table 1.1 Findings of sentiment analysis using CNN

Input: Output
{ [
"sentence": "I love my new phone " {
} "negative": 0,
"neutral": 0.417,
"positive": 0.583,
"sentence": "I love my new phone "
}
]

{ [
"Sentence": "Although I'm well in math but I don't {
Like math... " "negative": 0.238,
} "neutral": 0.624,
"positive": 0.138,
"Sentence": "Although I'm well in math but I don't
Like math... "
}
]

{ [
"Sentence": "First half is interesting, but last half is {
boring :(" "negative": 0.426,
} "neutral": 0.574,
"positive": 0,
"Sentence": "First half is interesting, but last half is
boring :("
}
]

neural network but the main obstacle of using CNN in this circumstance is it never
follows any concept of sparse tree for further proceedings whereas RNTN (Recursive
tensor Network) is concerned with the use of specific sparse tree for every input
data. In addition, the basic infrastructure in this work compute compositional vector
representation of every phrase in each node of a binary sparse tree constructed while
taking input. Further steps are followed by computation of parent vector in a bottom
up fashion corresponding to the input sparse binary tree.
Thus, the above Fig. 1.6 mentioned here represents the structured illustration
of input for recursive neural network. From the illustration we get informed that
hierarchical analyzing technique has been the main mechanism for this specific cir-
cumstance. For this contribution this method contributes to analyze the data collected
from one of the famous social media which is Facebook. Thousands of comments
along with various emotions, reactions, observations have been depicted here in
details. While working with bottom up fashion this algorithm basically labels all the
words through vector analysis using the following formula,

y a  so f t max(Ws a) (1.13)

For illustration of this equation we can conclude that Ws is the sentiment classification
matrix applying the overall proposed approach on which we will acquire our desired
output. Moreover, a is one of the vector through which every word has been labeled.
1 Teenagers Sentiment Analysis from Social Network Data 15

Fig. 1.6 Basic infrastructure

of a hierarchical sparse tree

However computing parent vector is mandatory for recursive neural tensor network.
Therefore, the computation of target vectors can be defined through the formula
mentioned below,
     
b a
t1  f W & t2  f W (1.14)
c t1

Here, f is considered as one of the element from standard nonlinearity function.

Also w is the main parameter to learn whereas bias has been omitted for simplicity
purpose. However, the bias can be added to the overall vectors through the addition
of an extra column especially for the bias. One of the main problem for almost all the
neural nets are seems to be appeared when the parameters become large and inhabited
as indivisible. For those circumstance recursive neural tensor network outperforms
most. Though the use of single layered vector form if we want to define overall output
the then that should be taken place as the following,
 T  
b b
Oi  Vi (1.15)
c c

For the context of Eq. 1.15 the V i is the tensor form of the multiple bilinear form.
Oi (Output) is the ith output tensor term which is the desired expectation of our
proposed system. For recursive neural tensor network to achieve any concluding
point the parent vectors usually defined as P1 &P2 must be defined and calculated.
For this contribution the parent vectors have been calculated through the use of
following formulation,
 T    
b i a a
Pi  f V +W (1.16)
c Pi Pi

Here, ith number if parents definition has been calculated through Eq. 1.16. Intu-
itively, one special case for the RNTN algorithm is when i  0 the infrastructure can
16 L. Rahman et al.

Fig. 1.7 Graphical representation of constituency based parse tree

directly relate to the input datasets which results in efficient and faster calculation to
the way of output.
For organizing sentences, recursive neural nets work as parsers, which after pars-
ing groups all the negative as well as positive words within particular groups for
further calculation. Some sub groups and phases are also created in this stage of
recursive neural nets. The sub phases includes Noun phases (NU), Verb phases (VP)
etc. further steps depend on various machine learning algorithms to classify exact
sentiment. By parsing the sentences, they can be successfully structured as tress
which is ordinarily known as parse trees (Fig. 1.7).
In [54], authors propose a Recursive Neural Tensor Network (RNTN). The contrib-
utors of this work draw their main concentration towards tensor-based composition
function for sentiment analysis. And record that RNTN perform better than other
methods of Deep Learning.
So we can use the idea of RNTN to analyze the sentiment. RNTN use 5 sentiment
classes, very negative to very positive (– –, –, 0, +, + +), at every node of a parse tree
and capturing the negation and its scope in given sentence.
In Fig. 1.8 we can see that RNTN cannot detect any emotion sign “”. And this
is like supervised learning so ‘boring’ get the very negative sign. Despite of these,
RNTN is quite good for text and its accuracy rate also good.

1.3.3 Detection of Positive, Negative and Neutral Sentiment

A simplest technique called ‘Keyword Spotting’ used in sentimental algorithms.

After input the data it is scanned for detecting the positive and negative words like
‘love’, ‘happy’, ‘good’, ‘bad’, ‘sad’, ‘terrible’ and ‘great’ etc. This approach is called
‘Bag of Words’ (BOW). This contains information of sentiment along with sentiment
scores. Like the word ‘bad’ is negative and the word ‘love’ is positive. Some list of
words in BOW given below in the Table 1.2:
1 Teenagers Sentiment Analysis from Social Network Data 17

Fig. 1.8 Analyzing a sentiment using RNTN algorithm

Then use negation detection measures to differentiate between ‘happy’ and ‘not
happy’. Thus we can go for a phrase ‘not happy’ is negative sentimental. Based on
those words, algorithm shows the sentiment rate of positive, negative and neutral.
Thus we do not use NLP, sarcasm problem might arise. For example, “Wow, thanks
for show your reality”, may be a negative sentence rather than positive one. To solve
it, we have to do Natural Language Processing. In our future work we will do this.
18 L. Rahman et al.

Table 1.2 List of positive Positive words Negative words

negative word from BOW
Happy Bad
Good Sad
Honest Cheating
Pleasant Annoying
Interesting Hate
Wonderful Sorrow
Love Evil
Like Dishonest
Gorgeous Impure
Fantastic Unhappy
Beautiful Ugly

1.3.4 RNTN Compared to CNN

By applying recursively same set of weights RNTN created. RNTN is successful

for tree structures, learning sequence in NLP. Sentiment analysis is possible through
the use of convolutional neural network but the main obstacle of using CNN in this
circumstance is it never follow any concept of sparse tree for further proceedings
where as RNTN (Recursive Neural Tensor Network) is concerned with the use of
specific sparse tree for every input data. For example, this is a movie review “First
half is interesting, but last half is boring”. The sentences have both positive and
negative sentiment. Thus RNTN create a tree structure where RNTN make two part
of this sentence based on some logical terms [56], one is positive part (First half
is interesting,) another is negative part (but last half is boring). But in CNN it
will show an overall rate of positive, negative and neutral based on words (positive,
negative).

1.3.5 Comparison of Different Sentimental Analysis

Techniques

In research area, sentiment analysis played an important rule. Many researchers

discussed about different methods of sentiment analysis. Many researches are still
going on to find out the efficient and alternatives approach for sentiment analysis.
There are many methods of sentiment analysis such as- Machine Learning Approach,
Rule Based Approach, Lexicon Based Approach, and others. Table 1.3. Shows three
main techniques used for sentiment analysis. Every technique has some strength and
limitation of its own.
Other documents randomly have
different content
 'But it 'll do us good if the machine crashes. I 'd be glad to know
it's out of action. Come on!'
They turned in the direction in which the sound was last heard.
Through the close-growing trees it was impossible to see far, and
Trentham privately thought the search a waste of time; but after
only a few minutes' walk they came to the edge of an open space
sloping down to a stream some twenty feet wide.
'It will be hereabout,' said Hoole, detaining the others at the top
of the slope. 'But I guess this trickle isn't wide enough to float it. Let
us separate, and scout along the line of bushes here, up and down
stream.'
In a few minutes Flanso, who had gone northward, returned to
the others, and told them by signs that he had discovered the
machine. Creeping back with him, they came to a bend in the
stream, and there discovered the seaplane, resting partly on two
small trees, partly on a bed of rushes, and awkwardly tilted. The two
airmen had left their seats, and were talking together on the bank,
apparently consulting a compass. Every now and again they glanced
apprehensively into the bush on both sides. Then they returned to
the seaplane, walked round it, put their shoulders against the
fuselage, and tried to lift it. One of them took out his revolver, and
was on the point of firing it into the air, when his companion hastily
interposed. The two men had a brief altercation. Finally the one who
had been about to fire appeared to yield to the other's warning, and
they both sat down on the shelving bank, discussing the position
over again.
Sheltered by the dense vegetation above the watercourse,
Trentham and his companions had watched their movements with
interest. The tenor of their discussion was easily divined. The
seaplane could not be salved without help, but they hesitated to
leave it, fearful of its being discovered by the natives, with whom, as
they now knew, were white men. If they parted company, which was
to return to the cove? The one left would be less able to defend
himself and the machine. A revolver shot might have brought
assistance from the Germans; on the other hand, it might attract a
horde of cannibals. What were they to do?
As they sat on the bank, they were sideways to the three men
watching them only a few feet above.
'Let's rush them!' whispered Hoole suddenly.
He seized Flanso's spear, pushed his revolver into Trentham's
hand, and before the latter could utter a word, either in assent or in
remonstrance, the American was half-way down the slope. Trentham
had no choice but to back him up, and he dashed after his friend
with scarcely a moment's delay.
The Germans heard the sound of Hoole's movements through
the bush, turned their heads and sprang up. One of them raised his
revolver to fire, but Hoole, now only three or four yards away,
launched his spear. His sudden action flurried the German's aim, his
shot flew wide, and the next moment he fell back, cursing, and
tearing the spear from his shoulder. His companion, seeing Trentham
rushing at him with levelled revolver, hesitated a moment, and
caught sight of Flanso swooping down immediately behind the
Englishman.
 ONE OF THE GERMANS RAISED HIS
REVOLVER, BUT BEFORE HE COULD
FIRE, HOOLE LAUNCHED THE SPEAR AT
HIM.

'Hands up!' cried Trentham, taking advantage of the man's

momentary hesitation.
Up went his hands.
'Guess we 'll borrow your revolvers, gentlemen,' said Hoole,
picking up the weapon dropped by the wounded man. Trentham
took the other man's from his belt. 'Keep your eye on them,
Trentham,' Hoole went on, 'while I kind of size up this machine of
theirs.'
Trentham and Flanso stood guard over the Germans while the
unwounded man bathed his comrade's arm and bound up his
wound. Meanwhile Hoole examined the seaplane in a manner that
took Trentham by surprise. There was a sureness, a purposefulness
about him; he seemed to know exactly what he was looking for.
Indeed, he pulled the engine about, as Trentham afterwards told
him, as if he were its maker. A very few minutes' inspection sufficed
to make him wise, as he put it.
'Not much wrong,' he said, coming over to Trentham and
smiling. 'I guess I can put it right. But we 'll want help to get it on to
the stream--yonder there, where it widens. Shall we start for home?'
'And these gentlemen will come as our prisoners?'
'Sure. We haven't any coal for them to dig, but they can start on
yams.'
'Ve are officers; it is not correct for officers to vork,' said one of
the Germans.
'Say, is that so? You 're a lazy lot? Well now! And yet you 'll
make a chief dig coal for you--a chief who 's as big a man here as
your Kaiser in Berlin. Well, you surprise me! Come along, Trentham.
Let's hurry.'
'How far are we from the coast?' asked Trentham of the
Germans as they started.
'Eight or nine mile,' was the surly response.
'Bully!' exclaimed Hoole. 'With luck we 'll have time to salve the
machine before it's found. Step along, Flanso!'
'Ze niggers vere ve go--are zey cannibals?' asked one of the
Germans anxiously.
'Yes,' replied Trentham. 'They nearly ate us. They mistook us for
Germans.'
The prisoners asked no more questions.
 Soon after leaving the seaplane, Hoole pointed out why it had
come down in this part of the forest. The stream widened into a
small lake, on which, when their engine failed, the Germans had
tried to alight. Unable to reach it, they had been forced to come
down on the bank of the stream.
Flanso scouted ahead, every now and then stopping to listen for
signs of the Germans. Once, when they were rounding a spur where
the vegetation was thin, Trentham clapped his hand over the mouth
of one of the Germans just as he was about to shout.
'We 'll have to gag this fellow, Hoole,' he said.
'Sure. Another rag from my coat. And look you here, you
officers, if you make a sound, barring a natural grunt, we 'll leave
you to our friend Flanso. See?'
'The native yonder,' explained Trentham. 'You had him on your
ledge, you know.'
The threat was enough. For the rest of the march the Germans
were docility itself.

CHAPTER XIV
AN INTERLUDE

At the entrance to the village the returning party found Grinson

alone, standing in the middle of the path, his knife in his right hand,
a spear in his left, and a dozen other spears on the ground beside
him. Neither Meek nor any of the natives was to be seen.
'What's the meaning of this, Grinson?' asked Trentham.
 'Ignorance, sir--just ignorance, poor heathens!' replied the
boatswain, cocking his thumb towards the village. 'That there airy
plane sailed over a while ago, and the savages all did a bunk,
screeching like one o'clock, though I roared myself black in the face
telling 'em 'twas only a sort o' bird. 'Twas no good; like as I 've seed
cows and sheep bolt with their tails up from a railway train. They
was scared stiff, there 's no mistake; and I 'spect they 're hiding
their heads somewhere.'
'And Meek?'
'I sent Ephraim straight into our hut, sir, and seeing as how the
whole place was left undefended, like, I took up my station here.
Germans, sir?' he added in a whisper, winking towards the prisoners.
'Yes; their machine came down. We 'll leave them in your
charge, as we want to take a gang back to bring in the machine. You
had no trouble on the way home?'
'Not a bit, sir, except that they 've no notion whatever of the
proper way of marching--more like a lot of colts they was; but there,
I hadn't the heart to correct 'em, they was so uncommon pleased to
be free again.'
The party had not interrupted their march. Grinson had picked
up his spears and fallen in step beside Trentham. As they passed
along the path, from behind the huts and the midst of the
plantations native heads appeared one by one, and when the
timorous people recognised their young chief they came bounding
out with yells of delight, until by the time the inner enclosure was
reached the whole population had joined the procession. At the gate
the patricians were assembled, headed by the temporary chief. They
welcomed Flanso with some show of dignity, and conducted him to
the chief's house, bowing low as he entered the doorway. The two
prisoners were placed under guard in an empty hut, and then
Trentham hurried after Flanso, and with the aid of the interpreter
explained the course of action which had been arranged between
himself and Hoole during the homeward march.
Trentham found himself contending with the natives' absolute
incomprehension of the value of time. Flanso was already occupied
in discussing with his elders the details of his approaching
installation. No western monarch could have been more deeply
absorbed in the ceremonies that were to inaugurate his reign. The
hardships from which he had been rescued, the dangers that still
threatened him and his people, seemed to have vanished from his
mind, and it was only by dint of patience and pertinacity that
Trentham succeeded in capturing his attention.
He pointed out that the Germans, enraged at the loss of their
prisoners, would certainly seek to regain them, and also to wreak
vengeance upon the community; nor would their animus be lessened
when they discovered that their airmen had fallen into the enemy's
hands.
'Chief say Toitsche fella no belongina find out that,' said the
interpreter, after an interruption from Flanso. 'Black fella belongina
eat white fella chop-chop.'
'Tell chief what he say all belongina gammon,' cried Trentham,
and proceeded to explain as well as he could from the slender
resources of pidgin English that the Germans might be valuable at
least as hostages. It took some time to get this theory understood
and accepted; then it was an equally long and difficult task to
persuade him that the seaplane would be of any value to him. What
was the good of it? It only frightened his people. To fetch it involved
the risk of falling into the power of the Germans. Trentham managed
to make him understand that the loss of the machine would greatly
cripple the enemy's operations; and further, that if Hoole succeeded
in repairing it, it might be used to bring help from friendly white
men, who would eat up the Germans, and deliver the natives for
ever from them. Flanso was rather impressed by these statements,
though he said that his people would probably prefer to eat up all
the Germans themselves; and Trentham realised the danger of
employing metaphorical language. Ultimately he brought Flanso to
concede his request--to despatch a party of able-bodied men to
transport the seaplane from its present position to the village.
'I feel utterly done up,' he said, mopping his brow, when he
returned to the others. 'Jabbering pidgin English for an hour is worse
than penal servitude. And it's such frightful loss of time; the
Germans may have discovered the machine by now.'
'Don't worry,' said Hoole. 'It was flying so low that I guess they
couldn't have seen it come down, and when they miss it they may
hunt for it for a day or two in the forest and not find it, except by a
fluke. I don't figure out that they 'll have all the luck. Anyway,
choose your men, and I 'll take 'em out; you 've done your share of
the business. I 'll take Grinson, he 's a hefty man, and may have a
notion or two.'
Fortunately the chief's obstinacy had no counterpart among the
Papuans outside the enclosure. The released prisoners had done
nothing since their return except relate over and over again the
details of their sufferings and the manner of their escape. Their
friends listened awestruck to the tale, and gasped as they heard of
the dangling rope, the lightning which had gleamed upon the cliff,
the crackling thunder, the strange stones that flew singing through
the air; and they looked with wondering admiration upon the white
men who had saved their fellows, not only from the tyrants who had
enslaved them, but even from the powers of nature which those
tyrants had at command. Trentham and Hoole thus found
themselves to be regarded with veneration, and when the
interpreter, prompted by Trentham, explained that the white men
required the services of twenty strong men to bring in the great sea-
bird which was another part of the enemy's magic, there was no lack
of volunteers eager to undertake the work. Another score were
selected as scouts, and when these understood that the object to be
carried was bulky, and could not easily be conveyed through thick
forest, they announced that they knew a way less obstructed by
vegetation, which would be more convenient, though less direct.
Under their guidance the party reached the stream some
distance above the spot where the seaplane lay. Feeling their way
cautiously along the bank, they came at length in sight of the
machine, which to all appearance remained exactly as it had been
left. Hoole took the precaution to post a screen of scouts around the
position to give warning if the enemy should approach, then he sent
Grinson to detach the wings. A handy man, like all British seamen,
Grinson soon accomplished his task, with the aid of tools discovered
among the airmen's outfit. Within a shorter space of time than Hoole
had deemed possible the work was finished. The wings were
entrusted to two men apiece; the body of the machine was hoisted
on the shoulders of the rest of the party; and although they met
with considerable difficulties at rough and steep places on the return
journey, once being saved from catastrophe only by the succour of
Grinson's sturdy muscles, they bore their burden without mishap to
the village, and carried it up the central path amid the joyous shouts
of the populace. Some of the men, now that the strange bird was
evidently helpless, showed their bravery by casting their spears at it,
and their dexterity by failing to hit any of the bearers. At this Hoole
fairly lost his temper, and rushed among the throng, smiting them
right and left with his fists. This unusual mode of correction was
effectual. The men who were sent spinning picked themselves up
with an air of surprise, while their comrades shouted with laughter,
in which the culprits themselves by and by joined. For safety's sake
Hoole had the machine carried into the inner enclosure, where it was
inspected with more decorum and shyness by the patricians, and
with contempt by the medicine-man, who demonstrated his
assurance by stroking the petrol tank and afterwards licking his
greasy fingers. The grimaces he made were so much like those he
was accustomed to display for professional purposes that his
discomfort passed unnoticed except by the white men.
Trentham came out of his hut rubbing his eyes.
'Yes, I 've been asleep,' he said, in answer to Hoole's inquiring
look. 'A pretty warden of the camp I should make. But the fact is,
these people are hopeless. I tried to make them understand that the
Germans might be upon us at any minute--no good! They appear to
be entirely taken up with some sort of mumbo-jumbo, and can't
attend to anything else. So I simply gave in, trusting that if the
Germans did appear the people would be scared into
reasonableness. The wall, of course, is proof against anything less
than a four-pounder.'
 'Well, I guess you did right,' said Hoole, 'and after thirty odd
hours without it, a little sleep would comfort me some. As to the
Germans, I 'm pretty sure they 'll do a bit of reconnoitring before
they attack. The surprise of those two airmen wasn't put on; it's
clear that Hahn said nothing about us, and they 'll be wondering
how many there are of us. And so, my son, we must persuade
Flanso to keep some scouts out with their eyes lifting. With proper
notice we could put up some sort of defence. But I hope we 'll get
away before it comes to that.'
'You can repair the machine, then?'
'I reckon I can, if I can get hold of a forge. But I 'm dead tired,
so I 'll turn in, if you 'll keep your eyes open a while. So long!'
That evening, as the four men sat together once more in their
hut, the two younger drew from Meek the story which he had
already related to Grinson on the way home. It was a very colourless
narrative--a recital of the cold facts in the fewest possible words,
without a touch of passion or indignation. Grinson, however, was not
the man to leave his mate's story unadorned.
'He 's an 'ero, gentlemen, that's what Ephraim is!' he declared.
'If ever he gets back to the old country, I lay the name of Ephraim
Meek--ay, and his picter too--will be in all the newspapers. I 'll see to
that. And the cinemas too; by gosh, I hear 'em now, the cheers of
the little kiddies and the sobs of the women and gals when they see
Ephraim tied up, like that chap as defied the lightning, bidding of
them German ruffians to do their worst; he 'd never dig coal for
them, not him! P'r'aps one of you young gentlemen will make a
pome out of it, like that one about "the boy stood on the burning
deck," you remember, or one I used to know years ago when I went
to school, about a British Tommy. I don't rightly recollect it, but 'twas
a Tommy in some heathen land as wouldn't bob his head to an idol,
or thing of the sort, though they killed him for it. 'Twud be a shame
if Ephraim wasn't put into some pome too--an 'ero like him!'
'I ain't got the figger of an 'ero, Mr. Grinson,' said Meek. 'Now if
it was you--a-going off singing to be eat--that 'ud made a picter. I
couldn't sing if 'twas me--I 'm sure I couldn't.'
'Why, that was only like a sheep bleating on the way to the
slaughter-house--'eroes don't baa. Ain't I right, gentlemen?'
'What do you say, Hoole?' said Trentham, feeling somewhat at a
loss.
'Well,' drawled Hoole, 'I guess heroes ain't cheap, anyway, and I
'm proud to know two, that's sure.'

CHAPTER XV
DUK-DUK

Trentham took turns with Hoole and Grinson to keep watch through
the night, leaving Meek to the recuperative force of sleep. No
untoward incident disturbed the hours of darkness, but there was a
good deal of noise in the village, the men chattering incessantly.
'Reminds me of the meetings of our old Urban District Council,'
remarked Grinson once, when Hoole relieved him. 'Fust one, then
another, then all together--and nothing settled after all.'
'I guess they 're fixing something,' replied Hoole. 'In my country
there 's a good deal of clack when we elect a new president. It's a
new chief here, you know.'
'Which it means a coronation, p'r'aps, or a beano of some sort,
sir. Well, we 'll see, if we live long enough. Good night, sir.'
Hoole had taken the precaution to have the seaplane placed
near the hut. In the morning, as soon as it was light, he made a
more thorough examination of the machine than had been possible
before.
'What precisely is wrong with it?' asked Trentham.
'Luckily not much,' said Hoole. 'These Huns are no sports or
they 'd have risked the few miles back to the cove. A couple of flying
wires are broken--that 's nothing--but see here, the water jacket
round this cylinder is cracked; the water (for cooling purposes, of
course) will all run out, and cause the engine to overheat. That's
why they came down, though, as I said, I 'd have risked it; all the
same, it wouldn't be safe to risk a flight of any length, and the thing
must be repaired. The mischief is, we 've got no solder, and you
can't mend a crack in metal without that, or something equivalent.'
'But you said, when you first saw it, that it could be repaired.'
'Well, yes, I did; forgetting for the moment where we are. You
haven't seen anything in the shape of a forge anywhere around,
have you?'
'No, but I suppose they must have one, or they couldn't make
their spears. I 'll ask the interpreter.'
But the word forge was unknown to the interpreter, and
Trentham's effort to explain in pantomime, by blowing an imaginary
pair of bellows, proved fruitless.
'Well, I 'll go and look around,' said Hoole. 'And meanwhile, old
son, don't you think you 'd better persuade the chief to send out
some scouts? If the Germans do have a notion to attack us, five
minutes' warning would give us time to get every one inside the
wall.'
'I 'll try. He may be more amiable than he was last night.'
Hoole perambulated the village for some time before he
discovered what he sought. At last, however, within a narrow
enclosure behind the huts, he noticed a young man sitting on a
frame-like chair, and vigorously working two long sticks up and
down. A low fence prevented Hoole from seeing the object of these
energetic movements, but a little smoke and an acrid smell like that
which comes from a blacksmith's forge drew him nearer, and looking
over the fence, his eyes gleamed at what he saw.
In front of the native stood two long cylinders of bamboo, about
three feet high. From a hole in the base of each ran a thin bamboo
pipe; the two pipes converged and met at a small heap of glowing
charcoal, which burnt more brightly with each thrust of the sticks as
the native worked them alternately up and down in the cylinders.
Hoole jumped over the fence, and eagerly examined this primitive
forge. At the lower end of each of the sticks was fastened a huge
bunch of feathers, resembling a mop; and these, pumped up and
down, caused a considerable draught, by means of which the smith
blew his fire to a heat sufficient to soften iron.
'Eureka!' cried Hoole, exulting.
He dashed back to the chief's enclosure, got his permission to
have the forge brought within the wall, and in half an hour was
busily engaged in the preliminary work of repairing the water jacket.
'It will be done by this evening,' he explained to Trentham. 'To-
morrow I 'll be off to Wilhelmshafen or any other old place where I
can find a white man, and then----'
'There 's enough petrol?'
'Enough for the flight out; I 'll get more when I land. Say,
though, we shall have to take the machine to the sea. She can't run
off, has only floats. That's a pity--waste of time, not to speak of the
risk of coming up against the Germans.'
'But there 's an opportunity of getting the machine carried to
the sea without a special journey. I 've heard some news while you
were away. It appears that some sort of ceremony inaugurating the
new chief is to take place at our old wreck to-night. All these
Frenchified men are going with him in procession, with a certain
number of the other fellows. We must get him to let his men carry
the 'plane at the same time.'
'They go down the chimney?'
'Yes; it's the nearest way.'
'The narrow way that leads to destruction! How on earth are we
to get the machine down there?'
'With ropes, man. We 've tested the quality of their ropes
already, and the women work so uncommonly fast that they 'll have
new ropes the right length in plenty of time. I 'll go and see the chief
about it at once. He 's very much preoccupied, and vastly self-
important, but he allowed me to send out scouts, as you suggested,
and I dare say I can talk him over.'
Flanso was quite willing that the seaplane should form part of
his procession. His vanity appeared to be flattered; he was as much
pleased as a Lord Mayor of London who has secured some novelty
for his show. But when the carriers had been chosen, an unforeseen
difficulty arose. The medicine-man, whose office gave him access to
the chief at all times, strutted into Flanso's house, where the elders
of the community were discussing the details of the approaching
ceremony, and vigorously protested against the seaplane being
allowed to leave the village. In a vehement oration he declared that
the strange bird must have some connection with the totem of the
tribe, and that while it remained with them the village would be safe
from hostile attack. Some of the elders backed him up, and Flanso,
torn between his own superstitions and his sense of loyalty to the
white men who had rescued him, sought relief from his perplexities
by sending for Trentham, and putting the case before him.
Trentham had sufficient diplomacy to conceal his amusement,
and also a certain irritation at the threat to his scheme.
'Tell chief,' he said to the interpreter, 'medicine-man fella he
savvy lot. Big bird belongina totem all right; all same big bird he fly
long way, bring back lot of white fella; they fight bad white fella this
side, eat bad white fella all up.'
At this moment, unluckily, one of the scouts returned with the
report that the big ship no longer lay in the cove, and that all the
white fellas had disappeared. Grinning with triumph, the medicine-
man instantly claimed that this fact proved his case; the loss of the
big bird had evidently rendered the enemy helpless, and there was
nothing further to be feared from them. Trentham, surprised as he
was at the departure of the Raider, and suspecting that the Germans
were probably setting a trap, strained his vocabulary of pidgin to the
utmost to counteract the medicine-man's arguments, and ultimately
prevailed on the chief to abide by the promise he had given. The
medicine-man and his supporters were patently annoyed. They left
the hut in undisguised ill-humour, and Trentham had an uneasy
feeling that they would still give trouble.
The procession was to start soon after sunset, so that it might
reach the wreck in time for the ceremony to take place at the height
of the moon. It was late in the afternoon before Hoole had
completed his repairs, and after making a good meal the four white
men were sitting in their tent, awaiting the moment for starting.
'What's up, Trentham?' asked Hoole. 'You look very sick.'
'I 've been thinking we 're mugs, that's all,' said Trentham. 'With
the chief and all his fighting men away, and us too, what defence
has the village if the Germans take it into their heads to attack?'
'Gee! I 've been so busy that I hadn't given it a thought. But the
Germans are out for slaves; they won't find any able-bodied men
here.'
'That's true; but you saw what they did to that village in the
forest. They 're capable of burning the whole place down, and
shooting the women and children, from sheer revenge and spite.
Hadn't we better wait till the chief returns before we start on our
own scheme?'
For a few moments Hoole pondered in silence.
'The Raider has gone,' he said at length. 'This means that there
can't be many Germans left behind; for, of course, they haven't all
gone. If the natives didn't see any, it's because they 're still
searching in the woods for the seaplane. Is it likely that the few left
will attack? There is a risk, I admit, and my proposition is that you
remain here with Meek and give an eye to things, while I take
Grinson to fix on the planes and see me off. The sooner I get help
the better, and the opportunity seems too good to be missed. What
do you say?'
'Very well. I should have liked to see the ceremony, but--we
simply can't leave the most helpless of the people to the Huns'
tender mercies. What in the world is this?'
At this exclamation the others, whose backs were towards the
gate, turned about. Through the gate was filing an extraordinary
procession.
'By cripes!' said Grinson. 'Is it Jack in the Green, or Guy Fawkes
Guy?'
A line of a dozen uncouth figures was slowly approaching.
Above legs bared to the thighs bulged a mass of leaves nine or ten
feet in circumference, crowned by a headdress like a candle
extinguisher, from which rose a pole, fantastically coloured, four or
five feet in height, with tufts of feathers and leaves at the top.
Except the legs, no part of the human form could be seen.
These strange figures came slowly across the enclosure, until
they reached the seaplane, the chief and all his men watching them
in dead silence. Encircling the machine, they stooped until the leaves
touched the ground; then, at a harsh cry from their leader, they
leapt into the air and began to dance; their leafy clothing rustled;
the poles wobbled and swayed; their legs bent and straightened;
and as they swung round and round the seaplane they uttered shrill
cries ever increasing in intensity. The white men looked on in
amazement. What was the meaning of these antics? Flanso and his
men seemed to be impressed. Trentham beckoned to the interpreter.
'What all this?' he asked.
'This duk-duk,' replied the man in an awed whisper.
 'Duk-duk! What is duk-duk?'
'Big medicine. Duk-duk dance; that say big bird no go away;
white fella all belongina afraid.'
Hoole whistled.
'That's the stunt, is it? By gosh, Trentham, it must be that old
medicine-man thinking he 'll scare us stiff. He 's gotten a kind of
affection for the machine. Well, Grinson, come along with me.'
He rose slowly, walked towards the seaplane, dodging between
two of the dancers, and got into the seat behind the engine. Grinson
had followed him.
'Just go to the propeller, Grinson,' he said, 'and swing it round
five or six times when I give you the wink.'
The dancers had drawn closer to the machine, yelling more
shrilly than ever. Hoole watched them with a smile as they circled
round. Suddenly he gave the sign; Grinson caught the propeller, and
with a heave of his brawny arm swung it about. Hoole nodded to
him to step aside. The leader of the dance was just approaching,
when there was a roar; Hoole had started the engine, and the
propeller whizzed round with ever-increasing velocity. The dancer
stopped short; before he could collect himself or retreat the air set in
motion by the whirring propeller smote him with hurricane force,
stripping the leaves from his body, and whirling his headdress, pole
and all, across the enclosure. There stood revealed the lean, naked
form of the medicine-man. He threw up his hands as if to defend his
face from the blast; then, with a yell of fury, he sprinted to the gate,
followed by the rest of the dancers.
 THE LEADER OF THE DANCERS WAS JUST
APPROACHING WHEN THERE WAS A
ROAR, AND THE WHIRRING PROPELLER
SET UP A HURRICANE WHICH CAUGHT AT
HIS DRESS.

A great roar of laughter burst from the spectators, hitherto

silent, and from the crowd which had gathered outside.
'I guess the duk-duk won't quack so loud in future!' said Hoole,
rejoining the others.
The medicine-man's attempt to retain the seaplane and
demonstrate his own importance had ignominiously failed. His
erstwhile supporters had no more to say. The carriers were called
up; the chief's procession was formed, and when the red moon
rested on the horizon they set forth solemnly towards the ancestral
wreck.

CHAPTER XVI
FLIGHT

'I don't think nothing of this 'ere percession, sir--nothing at all,' said
Grinson, tramping beside Hoole. 'It ain't my idea of a percession--not
by a long chalk. I 've seed hundreds of Lord Mayor's Shows, and
King George a-going to be crowned. They was somethink like--
everybody moving like clockwork, 'cept the horses, and they did
their best. But these 'ere cannibals ain't got a notion o' keeping step.
Look at 'em! What's the good of their drums, I 'd like to know? Why
don't they tap out the left-right, as you may say, so 's they will
march proper?'
'Well, I guess they 're children of nature,' said Hoole with a
smile.
'Brought up very bad, then, that's all I can say, sir. I 'm glad I 'm
not a child of nature, but of respectable parents, and uncles and
aunts the same. My old Aunt Maria, now, she was real respectable,
and no mistake. I used to go and see her when I was a nipper. Fust
thing she 'd do, she 'd wipe me nose. It ain't much of a nose--not
like Ephraim's--p'r'aps he didn't have his wiped so often, so it
growed better.... Why, save us, sir--did you hear that?'
 Both the men turned round. From the rear of the procession
came the cry of a white man in mortal peril. The moonlight, striking
along the leafy avenue through which they were passing, revealed
the hideously painted faces of the Papuans, some of whom were
carrying live pigs for the feast on the shore. Again rose the startling
cry. The two men, stepping among the trees to avoid hindering the
march, walked back quickly, and presently saw, among the black
faces, the pale, terrified features of the German prisoners, whom
rough hands were dragging along at the tail of the procession. An
explanation of their presence flashed upon Hoole. Baulked of their
human prey before, the cannibals had determined that this time the
custom of their tribe should be followed. The inauguration of their
chief should not pass without its human sacrifice.
'Hi, you blackguards! What about my goose?' cried Grinson, and
was rushing to the rescue when Hoole detained him.
'It's no use tackling these hobos,' said the latter. 'They 'd fly at
you like dogs. Keep pace with them; I 'll run on and talk to the chief.'
Hoole sprinted up the line, past the Papuans, carrying food and
cooking-pots, past the drummers and the men who bore the
seaplane, past spearmen and dancers, and the retinue of patricians
who followed immediately behind Flanso. Laying a hand on the
chief's arm, he remembered that the interpreter had been left in the
village, and wondered nervously how he was to make Flanso
understand. To speak in English would be useless; he knew but a
word or two of the native language. Suddenly he remembered the
chief's remote French extraction, and the impression made on his
father by Trentham's use of French. Could that be turned to
account? To explain in French what he desired was beyond him,
even if Flanso could understand a continuous speech; what
alternative was there? Perhaps if he could remember some French--a
French poem, say--and declaim it with appropriate gestures, it might
produce the moral effect which was the first necessity at the
moment. Hoole was a good elocutionist and amateur actor. Racking
his memory for things learnt at school, he could recall but one
poem, a song of Béranger's that had taken his fancy. He posted
himself at the head of the procession, and facing Flanso, spread his
arms to signify that no further advance was to be made. Impressed
by his determined mien, the natives halted. And then Hoole, raising
himself to his full height, began to recite:

'Il était un roi d'Yvetot,

Peu connu dans l'histoire.'

Rolling the r, he pointed dramatically to the moon,

'Se levant tard, se couchant tôt,

Dormant fort bien sans gloire.'

Here he shot out his right hand towards the village. The natives
gasped.

'Et couronné par Jeanneton

D'un simple bonnet de coton:
Oh! Oh! Oh! Oh! Ah! Ah! Ah! Ah!
Quel bon petit roi c'était là!
La, la!'
He accompanied the refrain with a rhythmical clapping of hands;
then, without waiting for the mystic effect to dissipate itself, he
grasped the chief by the arm, and led him back along the halted line
to the spot where the Germans stood paralysed in the hands of their
captors. French, he feared, would hardly serve him now; but he
pointed to the prisoners, and repeated 'Pas manger; pas manger!' in
a fiercely authoritative voice, winding up with 'En arrière!' and
pointing towards the village, from which they had come but a
quarter of a mile. Then, stepping firmly up to the group of Papuans,
he seized the arm of a big man who clutched one of the Germans,
and with a dexterous twist forced him to relax his hold. The men
snarled, but Flanso, who by this time had apprehended what the
white man was driving at, sternly ordered the prisoners to be
released.
'You must take them back, Grinson,' Hoole said. 'Their hands
are tied; I 'll fasten them to you so that they can't get away. Take
them back and hand them over to Mr. Trentham; then catch us up.
Bring Lafoa back with you.'
'Ay, ay, sir. I 'll take a spear, case of accident. And I won't be
long behind you.'
Hoole, sweating at every pore, as he afterwards said, gave an
off-hand indication of approval to Flanso, and the march was
resumed.
It was a slow and wearisome journey through the forest, and
Grinson, accompanied by the interpreter, caught up the procession
before it had advanced another mile. Many times the bearers of the
seaplane had to rest; in some places trees and shrubs had to be cut
down to allow its passage, and Hoole had many an anxious moment,
fearing that the wires would be bent or broken through the
carelessness of his inexperienced carriers. But the edge of the cliff
was reached without serious mishap, and in a much shorter time
than he had believed to be possible. The route chosen by Flanso was
in fact shorter by many miles than that by which Hahn had led the
party weeks before.
The covering over the top of the chimney had been removed,
and Flanso and his men nimbly climbed down one by one. Hoole
allowed all to descend except the carriers, Grinson, and the
interpreter; then he set to work to ensure the safe lowering of the
machine and its detached planes. With the interpreter's aid he got
the men to make a stout platform of saplings extending some feet
beyond the edge of the cliff, with a hole at the chimney top. The
body of the machine was secured to two stout ropes which were
carried round trees. Hoole had a third rope with a sling in which to
put his feet, and this he tied firmly to a trunk.
'Now, Grinson,' he said, 'you will be in charge here, and woe
betide you if you fail me! I 'm going to let myself down into the
chimney. When I give the word, lift the machine gently over the cliff
and lower away; I 'll keep an eye on it and fend it off the rocks.
That's all clear?'
'As clear as good beer, sir, and I only wish I had some, I 'm that
dry. Ephraim says there 's----'
'Oh, hang Ephraim! Stand by!'
He swung himself into the chimney and shouted 'Lower away.'
 'The next ten minutes,' he told Trentham afterwards, 'were an
age--an epoch. I heard Grinson yo-hoing to the natives above; he
never stopped for a moment. How he managed to control those
savages I don't know. Down came the machine, and there was I,
swinging on the rope, clinging with one hand while I guided the
machine with the other. Once it dropped suddenly on one side, and I
feared the whole caboodle would crash to the ground; but Grinson
bellowed like a hurricane, the thing righted itself, and my heart was
banging against my ribs like a steam hammer. Well, we got the
fuselage down at last; the planes were a simpler proposition, and by
and by Grinson joined me on the beach, as proud as a peacock. "I
done that in fust-rate style, sir," says he with a broad grin. "You can
bear me out, sir."'
Hoole was so much occupied in putting the machine together by
the light of the moon that he saw little of the native ceremony.
There was drumming and dancing; at one moment the chief,
followed by a group of his men, marched solemnly to the wreck, and
after tramping seven times round the broken mast, descended to the
cabin. On his return he was welcomed with a frenzied shout. Then
fires were lighted under the cooking-pots, the dance was kept up
until the viands were ready, and the feast was prolonged until the
moon had travelled half-way round the sky. Satiated, the natives
flung themselves on the sand and slept.
'Well, sir,' said Grinson, sitting beside Hoole, close under the
cliffs, 'all I can say is, it's a wonder to me how they could stow away
so much boiled pork without mustard or a pint of beer.'
Morning dawned. Hoole had the seaplane carried down to the
sea. The tide was rising; the sea was choppy; waves of considerable
size were breaking on the beach.
'I hope she 's seaworthy, sir,' said Grinson, eyeing the machine
anxiously.
'That we 'll soon discover,' replied Hoole. 'You know what you
have to do. When I 've pumped up and sucked in, you 'll give the
propeller half a dozen turns; then I 'll switch on, and you 'll skip
away, or you 'll get a dose like the medicine-man.'
The whole body of natives was lined up on the beach, watching
the white men with simple curiosity. Suddenly one of them gave a
shout, and pointed out to sea.
'By the powers, sir, here's the Blue Raider,' cried Grinson.
The vessel had appeared round the western horn of the bay,
and was standing in as close as possible to the shore.
'They must have seen us,' said Hoole. 'Rank bad luck! Think I 'd
pass for one of the Germans, Grinson? They 're too far off to
distinguish features.'
'You might, sir, though I 'm ashamed to say it; but what about
me? I 'm a good deal too broad in the beam.'
'But you 're half in the water. I 'll try making signs, just to keep
'em quiet, for if they 're suspicious and get their guns on to me, I 'm
a dead man, sure!'
Partly concealed by the overshadowing planes, Hoole waved his
arms in imitation of the Morse signals. From the vessel, which had
now hove to, a boat was lowered, and pulled towards the beach.
'Now, Grinson, I must cut,' said Hoole. 'I 'll start towards them;
that will diddle them, I hope; and before they make up their minds I
'm an enemy, I guess I 'll be out of range. Ready?'
 Grinson whirled the propeller; Hoole made contact and switched
on, the engine started with a roar, and the seaplane glided forward
in the direction of the approaching boat. Cries of wonder broke from
the throng of natives as the strange bird rose into the air. When
clear of the water it turned suddenly eastwards, and flew rapidly at a
low altitude along the coast. The boat stopped. A signal was fired
from the vessel in the offing. The seaplane continued on its
eastward course. There was a loud report which sent the natives
scurrying to the foot of the cliff, and a shell burst a few yards behind
the seaplane. A few moments later a shell dropped on the beach,
sending up a great shower of sand and rocks. The natives scattered,
some making for the chimney, others taking cover among the rocks
and undergrowth. Grinson stood for a while with arms akimbo at the
edge of the beach, as though defying the gunners; then a crackle of
rifle fire from the boat sent him too rushing for the chimney. He had
only just reached it when a shell fell plump into the wreck, scattering
its timbers far and wide. A howl of wrath and dismay burst from the
natives, and they began to swarm up the chimney, full in view of the
occupants of the boat, but concealed from the Raider by a bulge in
the cliff. Bullets rained on the rocks, but the rolling of the boat
rendered marksmanship difficult, and only one or two of the men
suffered slight wounds as they climbed up. Grinson was the last to
leave the beach. When he reached the summit he stood for a
moment or two on the platform, gazing with a grim smile at the
Raider.
'Blue you are, and well you may look it!' he cried. 'Row back,
you lubbers; such darned bad shooting I never did see, and I 'll take
my davy as your goose 'll soon be cooked. Mr. Hoole has gone to
fetch the sauce. Ahoy, ugly mugs! Wait for me! I don't want to be
marooned in this 'ere forest.'
And he set off at a trot to overtake the natives, who were
already disappearing among the trees.

CHAPTER XVII
THE ATTACK ON THE VILLAGE

In the haste of their flight from the dangerous beach the natives had
left drums, cooking-pots, and other impedimenta behind, and had
now nothing but their weapons to carry. Marching light, they showed
a most ungentlemanly want of consideration for Grinson. 'They 're
no gentlemen, Ephraim, me lad, that's flat,' he said later (they had
gone several miles before he overtook them), 'and me panting and
sweating like any think! I must 'a lost five pounds, if an ounce.'
As soon as he had recovered breath, he took place by Flanso's
side at the head of the procession, and quite unaware that he was
transgressing etiquette, kept pace with the chief, conversing by the
way.
'What I can't make out, Flanso, old man,' he said, 'is what this
Mr. Hoole is--what he was brought up to, like. He 's only a nipper, so
to speak it, 's far as years go, and by his own account he was just
larking round these 'ere seas, with more money than wits, I 'd say.
Well, then, what I want to know is, how does he know about these
'ere airyplanes or seaplanes, or whatever you call 'em, handles 'em
as easy as I might handle a ship's boat, and no gammon? I know
you don't understand a word of what I'm saying, matey, not being
born such; but I 'm just letting off steam, d' ye see? In course I
wouldn't say all this to the young gent's face, my manners being
good as a general rule; but it ain't good to keep your thoughts
bottled up, like, 'cos they might bust out sudden like the cork out of
a bottle o' beer that's stood too near the fire, and then where 'd we
be, Flanso, me boy?'
Flanso appeared to be gratified by the seaman's speech, and
smiled amiably.
'You ain't half a bad-looking chap, Flanso, and if you 'd take that
stick out of your nose and learn to wash yourself you 'd be quite
handsome. 'Tis a strange world, and no mistake. There 's my
Ephraim, a good lad as cleans hisself regular like a true Christian,
and speaks English as well as I does myself, but nobody could say
he 's a beauty; and then there 's you--'orrible dirty, speaks like a
monkey, and yet got a face on you as would make your fortune if so
be you went into the Guards.
'Things is arranged very rum, Flanso, and there 's no
understanding of 'em. Them two Germans, now--they 're pretty
enough with their blue eyes and flaxen hair, just the sort the girls go
silly over; but their 'earts, Flanso--their 'earts is blacker than you, old
man. Yet I don't know--p'r'aps--but what's the good o' spekylatin'?
Things is a puzzle, and we ain't got the brains for to work it out.'
He shook his head, and fell into a reverie from which he was
suddenly awakened by a distant rifle shot. Instantly he became the
man of action again. He swung round, signed to the natives to halt,
and put his fingers to his lips enjoining silence. Then he hastened
down the line until he had discovered Lafoa.
 'Here, you!' he said. 'You makee all fellas hold their jaw and
come bunko arter me, see? Bad fellas fightee; we go see what's up.
See?'
The interpreter nodded, and hurried to Flanso, who gave orders
in his own tongue. It was clear that he looked to Grinson for
leadership. More shots rang out ahead.
'That's a revolver,' murmured Grinson. 'Mr. Trentham! Spears.
Bows and arrows.' He looked round at the natives. 'May be some
good. Fust thing is to find out exactly what's what. Here, you, how
far off--how muchee way--your village?'
The interpreter did not understand him. Grinson groaned.
'Well, makee all black fellas hidee in bushes, so no one can see.
Savvy?'
This appeared clear, for the natives, at a word from Lafoa,
scattered, and in a few moments were invisible.
'Now come alonga me,' continued Grinson. 'Flanso too. Come
on!'
Flanso leading, they wormed their way stealthily through the
forest growth. The sounds of conflict grew as they progressed.
Presently they were conscious of the smell of burning, and thin trails
of smoke were wafted among the trees. Coming to the outskirts of
the village, they beheld several huts blazing. At first they saw no
human being, but advancing cautiously through the thick bushes
that spread behind the huts, they came upon a Papuan squatting on
the ground and rocking himself with pain. A question from Flanso
elicited an account of what had happened. An hour or two before,
the outlying scouts had discovered a number of white men marching
towards the village. They had hastened back with the news, and the
white men in the village had ordered all the people to withdraw into
the inner enclosure. The enemy had come upon them before they
were all within the wall. Some of the people had been killed, some
wounded, others had escaped into the forest. The huts had been
fired, and the enemy had rushed towards the gate, but had hurriedly
retreated before a shower of arrows and the fire-magic that met
them. They had posted themselves among the trees, and a little
while before had begun to shoot their hot stones over the wall.
Lafoa interpreted this as well as he could to Grinson, who,
however, seemed to understand the position instinctively. He bade
Lafoa hasten back to the men halting in the forest a quarter of a
mile behind, and bring up all the young warriors. While the man was
absent Grinson sat down at the foot of a tree, stretched out his legs,
folded his arms, and pondered. There had not been time for the
Raider to reach the cove and land her crew, therefore the attackers
must be relatively few in number. They were armed with rifles;
probably the rifles had bayonets; therefore, though few in number,
they were much more than a match for a throng of untrained
savages with no better weapons than bows and arrows and spears.
'Which I mean to say,' said Grinson to himself, 'that is if we
don't come to close quarters, whereas and however 'tis numbers
what tells in a rush, such as boarding a vessel when you 're close
alongside and Admiral Nelson piped all hands to repel boarders--or
t'other way about, for 'twas us what boarded the enemy, must 'a
been, being British. That's one point settled, Flanso, old man; we 've
got to board 'em, take 'em abaft, otherwise about the rump or stern,
and lively too, not forgetting that Mr. Trentham is for'ard banging
away with his revolver, and we stand a good chance of being bowled
over in mistake, and apologies are no use, and it's our look-out, and
no blame to anybody. Now I come to think of it, that point's fust and
last, for if we rush 'em and don't win, why, then nothing matters no
more, and we 're all booked for the pearly gates and no mistake.
Things are getting pretty hot by the sound of it, and I 'm afraid that
there revolver have give out, and--ah! here come the boarding
party.'
Lafoa stole through the brushwood followed by twenty or thirty
young men of the higher caste, all moving as silently as wild
animals.
'Tell 'em to lay snug--otherwise doggo--while you and me and
the chief go for'ard and do a bit of spying,' said Grinson.
He crept forward with the two men, and came presently to a
spot among the trees where it was possible to get a fairly clear view
of the fighting. A number of Germans were laying piles of
brushwood at the foot of the wall; others were breaking holes in it,
or enlarging slight gaps between the logs; others were bringing up
more brushwood from the forest, while reports and flashes high up
in some trees overlooking one side of the enclosure showed that
snipers had been posted to pick off the garrison. For the moment
none of the defenders were visible. Grinson guessed that Trentham,
having exhausted or all but exhausted his ammunition, was
reserving himself for the inrush which without firearms it was
impossible to prevent.
'There 's Trousers!' thought Grinson, catching sight of Hahn,
who appeared to be superintending operations. 'If I don't give him a
dusting----'
 He stole back to the waiting natives, whispering instructions to
Lafoa on the way. Half a dozen men were told off to creep through
the bush and deal with the snipers in the trees, Grinson judging that
in the dense undergrowth a native with a bow and a spear should be
a match for a white man in a tree, even though armed with a rifle.
The rest of the men were ordered to follow him noiselessly to the
edge of the clearing in front of the wall, and to dash at the enemy
when he gave the word. The slight sounds of their movements were
smothered by the reports of the rifles and the hacking at the wall.
On reaching the spot, Grinson perceived that the Germans were
massing for a determined push. At the sound of Hahn's whistle they
sprang on to the piles of brushwood and attempted to swarm up and
over the wall. Grinson could just see Trentham above the top,
swinging a huge native club. While he was looking, the pendulum
swing of the weapon disposed of two Germans who had come within
its formidable sweep. To right and left, however, several of the
enemy had got a footing on the wall, in spite of the spears hurled by
the older and weaker natives who alone had remained in the village.
Then Grinson let out a bellow like the blast of a fog-horn, and
sprang from the shelter of the trees with a spear in each hand,
followed by the horde of natives, yelling and screeching. The
Germans turned in alarm to discover what was threatening them,
dropped to their knees, and raised their rifles. Only a few of them
fired; the rest, disconcerted by the shower of arrows and spears
which the natives let fly at them as they ran, dropped their rifles and
fled helter-skelter among the trees, Flanso leading his men in hot
pursuit.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com