Full Download Social Media Analytics For User Behavior Modeling: A Task Heterogeneity Perspective 1st Edition Arun Reddy Nelakurthi PDF
Full Download Social Media Analytics For User Behavior Modeling: A Task Heterogeneity Perspective 1st Edition Arun Reddy Nelakurthi PDF
com
https://textbookfull.com/product/social-media-
analytics-for-user-behavior-modeling-a-task-
heterogeneity-perspective-1st-edition-arun-reddy-
nelakurthi/
https://textbookfull.com/product/social-computing-and-social-media-
user-experience-and-behavior-gabriele-meiselwitz/
textbookfull.com
https://textbookfull.com/product/network-intelligence-meets-user-
centered-social-media-networks-reda-alhajj/
textbookfull.com
https://textbookfull.com/product/naturalism-in-the-philosophy-of-
health-issues-and-implications-1st-edition-elodie-giroux-eds/
textbookfull.com
Origami Polyhedra Design 1st Edition John Montroll
https://textbookfull.com/product/origami-polyhedra-design-1st-edition-
john-montroll/
textbookfull.com
https://textbookfull.com/product/life-care-planning-and-case-
management-handbook-fourth-edition-berens/
textbookfull.com
https://textbookfull.com/product/it-s-not-you-identifying-and-healing-
from-narcissistic-people-durvasula-phd-ramani/
textbookfull.com
https://textbookfull.com/product/narratology-introduction-to-the-
theory-of-narrative-mieke-bal/
textbookfull.com
Social Media Analytics
for User Behavior
Modeling
A Task Heterogeneity Perspective
Data-Enabled Engineering Series
Series Editor: Nong Ye, Arizona State University, Phoenix, USA
Data Mining
Theories, Algorithms, and Examples
Nong Ye
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize
to copyright holders if permission to publish in this form has not been obtained. If any copyright material
has not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, trans-
mitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access
www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC),
222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that pro-
vides licenses and registration for a variety of users. For organizations that have been granted a photocopy
license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Preface.......................................................................................................................ix
Acknowledgment ......................................................................................................xi
Authors....................................................................................................................xiii
Contributors ............................................................................................................. xv
v
vi Contents
Chapter 6 Conclusion.....................................................................................81
6.1 User Behavior Modeling in Social Media ............................81
6.2 Addressing and Explaining Task Heterogeneity...................82
6.3 Limitations............................................................................83
6.3.1 Addressing Concept Drift ........................................83
6.3.2 Model Fairness.........................................................83
Contents vii
Bibliography ...........................................................................................................87
Index........................................................................................................................97
Preface
User-generated social media content provides an excellent opportunity to mine data
of interest and helps in developing functional data-driven applications. The rise in
the number of healthcare-related social media platforms and the volume of health-
care knowledge available online in the last decade have resulted in increased social
media usage for personal healthcare. In the United States, nearly ninety percent of
adults, in the age group 50-75, have used social media to seek and share health
information. Motivated by the growth of social media usage, this book focuses on
healthcare-related applications, studies various challenges posed by social media
data, and addresses them through novel and effective machine learning algorithms.
The content presented in this book will be of great interest to students and
researchers in the field of Machine Learning with applications to Social Media and
Healthcare. This book assumes the reader has sufficient understanding of the theory
of machine learning and linear algebra. The reader is suggested to refer to the relevant
literature cited for better understanding of proposed frameworks. We are also grate-
ful to collaborators, Dr. Curtiss B. Cook, Dr. Ross Maciejewski and Dr. Hanghang
Tong for their valuable feedback and suggestions. We also thank all the colleagues at
the STAR and DATA labs. Thanks to Dawei Zhou, Yao Zhou, Xu Liu, Xue Hu, Jun
Wu, Lecheng Zheng, Pei Yang, Liangyue Li, Chen Chen, Xing Su, Si Zhang, Boxin
Du, Qinghai Zhou, Jian Jian Kang, Zhe Xu, Scott Freitas, Haichao Yu, Ruiyue Peng,
Rongyu Lin and Xiaoyu Zhang for their support.
ix
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
Acknowledgment
This work is partially supported by the United States National Science Foundation
under Grant No. IIS-1552654, and Grant No. IIS-1813464, the U.S. Department of
Homeland Security under Grant Award Number 17STQAC00001-02-00, and an IBM
Faculty Award. The views and conclusions are those of the authors and should not
be interpreted as necessarily representing the official policies, either expressed or
implied, of the funding agencies or the government.
xi
Authors
Arun Reddy Nelakurthi is a senior engineer in Machine Learning Research at
Samsung Research America, Mountain View, California. He received his PhD in
Machine Learning from Arizona State University in 2019. His research focuses
on heterogeneous machine learning, transfer learning, user modeling and semi-
supervised learning, with applications in social network analysis, social media analy-
sis and healthcare informatics. He has served on the program committee for The Con-
ference on Information and Knowledge Management (CIKM) and The Pacific-Asia
Conference on Knowledge Discovery and Data Mining (PAKDD). He also worked
as a reviewer for IEEE Transactions on Knowledge and Data Engineering (TKDE),
Data Mining and Knowledge Discovery (DMKD) and IEEE Transactions on Neural
Networks and Learning Systems (TNNLS) journals.
xiii
Contributors
Curtiss B. Cook
Mayo Clinic
Scottsdale, Arizona
Ross Maciejewski
Arizona State University
Tempe, Arizona
Hanghang Tong
University of Illinois
Urbana-Champaign, Illinois
xv
1 Introduction
In recent years, social media has gained significant popularity and become an essen-
tial medium of communication. According to a survey, about 88% of the public in the
United States use some form of social media, a 53% growth in the last decade. Also,
the average number of accounts per user has increased from two in 2012 to seven
in 2016 [Pew Research Center, d]. The rise in social media usage both vertically in
terms of the number of users by platform and horizontally in terms of the number
of platforms per user has led to a data explosion. Popular social media platforms
like Facebook, Instagram and Twitter manage tens of petabytes of information with
daily data flows of hundreds of terabytes and a continually expanding userbase [Pew
Research Center, c]. Such huge volumes of user-generated content provide an excel-
lent opportunity to mine data of interest. We can, thus, look for valuable nuggets of
information by applying diverse search (information retrieval) and mining techniques
(data mining, text mining, web mining, opinion mining).
User-generated content is diverse based on the need the social media platform
caters to. Per one survey, amongst those who use social media roughly 67% stated
staying in touch with current friends and family as a major reason, while 17% felt
social media enabled them to connect with friends they have lost touch with [Pew
Research Center, a]. Other research indicated about 67% of the United States popu-
lation use social media to stay updated on the latest news and seniors are driving that
number up [Pew Research Center, b].
Social media usage has also seen a spike when it comes to personal healthcare.
In the United States, nearly 90% of adults, in the age group 50-75, have used social
media to seek and share health information [Tennant et al., 2015a]. Research demon-
strates that online social support programs like health care forums and social media
websites (e.g. Facebook and Twitter) can help patients gain knowledge about their
diseases and cope better with their daily management routine [Petrovski et al., 2015].
Effectively mining information from these healthcare-related social media plat-
forms can, thus, have a wide range of applications resulting in improved healthcare.
For example, healthcare social networks can connect patients suffering from major
chronic diseases such as diabetes mellitus, with physicians as well as other patients.
Compared to generic social networks such as Twitter and Facebook, disease-specific
social networks (e.g., TuDiabetes1 and DiabetesSisters2 ) have a greater concentra-
tion of patients with similar conditions and relevant resources. However, when it
comes to such social networks, the patient is more likely to stick to a single social
network, and would rarely look at other networks, thus limiting their access to online
resources, especially patients with similar questions and concerns. Identifying patient
1 http://www.tudiabetes.org/
2 https://diabetessisters.org/
1
2 Social Media Analytics for User Behavior Modeling
groups with similar conditions can help connect patients across networks, thereby
opening doors for knowledge sharing to help the community as a whole. Addition-
ally, in a world of “fake news”, a lot of health information is misrepresented and
therefore calls for authenticity. Motivated by the immense scope of leveraging social
media information for healthcare and addressing underlying challenges with usage
and reliability, in this book we explore answers to the following questions:
D1. Social media in healthcare: To study the real-world impact of social media
as a source to seek and offer support to patients with chronic health condi-
tions.
D2. Learning from task heterogeneity: To propose various models and algo-
rithms to learn and model user behaviors on social media platforms, to iden-
tify similar actors across social networks, to adapt and leverage information
from existing black-box models to improve classification accuracy under
domain adaptation settings.
Introduction 3
The book is organized as follows. The related work, Chapter 2 discusses existing
research and how the proposed methods differ from it. Chapter 3 discusses the impact
of social media on patients with diabetes mellitus. Chapter 4 presents algorithms and
models to learn from task heterogeneity in social media. Chapter 5 discusses methods
to explain task heterogeneity. Finally, Chapter 6 concludes our research.
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
2 Literature Survey
Since 2004, the growth of social media has been near exponential [We Are Social].
According to a survey, about 88% of the public in the United States use some form of
social media, a 53% growth in the last decade [Pew Research Center, d]. This growth
in social media usage led to an information explosion. Mining valuable nuggets of
data from such information generated through social media has immense applica-
tions [Zafarani et al., 2014]. Machine learning techniques have been widely adopted
to mine and analyze the large social media data to address many real-world problems.
Mining from social media platforms has many applications, (1) Event detection -
Social networks enable users to freely communicate with each other and share their
recent news, ongoing activities or views about different topics. As a result, they can
be seen as a potentially viable source of information to understand the current emerg-
ing topics/events [Nurwidyantoro and Winarko, 2013]; (2) Community detection -
identifying communities on social networks, how they evolve, and evaluating identi-
fied communities, often without ground truth [Zafarani et al., 2014]; (3) Recommen-
dation in social media - recommending friends or items on social media sites [Ricci
et al., 2011]; (4) Sentiment and opinion mining - identifying collectively subjective
information, e.g. positive and negative, from social media data [Liu, 2012]; (5) Net-
work embedding - assigning nodes in a network to low-dimensional representations
and effectively preserving the network structure [Cui et al., 2017].
As mentioned earlier in the introduction chapter, the intrinsic property of data het-
erogeneity in social media data poses a set of challenges. In this chapter, we present
the existing research on handling data heterogeneity and study the impact of social
media. In this chapter, we present existing work on impact of social media and its
implications in Section 2.1, Section 2.2 presents existing research addressing data
heterogeneity with a focus on task heterogeneity. Finally, we discuss the existing
research on explaining models under task heterogeneity in Section 2.3.
5
6 Social Media Analytics for User Behavior Modeling
YouTube has dramatically changed the way campaigns are run and how Americans
interact with their elected officials [Bonilla and Rosa, 2015]. Social media usage has
also seen a spike when it comes to personal healthcare. Tennant et al. [2015b] showed
that nearly 90% of adults who use the internet and social media platforms like Face-
book and Twitter used these platforms to find and share healthcare information. With
a lot of growing interest and immense benefits from healthcare applications to soci-
ety, we are motivated to work on addressing challenges in healthcare-related social
media platforms.
Research demonstrates that online social support programs like health care forums
and social media websites (e.g. Facebook and Twitter) can help patients gain knowl-
edge about their diseases and cope better with their daily management routine [Petro-
vski et al., 2015]. Patel et al. [2015] studied the impact of social networks on per-
ceived social support (e.g., of patients with chronic diseases). Researchers also stud-
ied how social media users gather and exchange health-related information and share
personal experiences [Naslund et al., 2016, Shepherd et al., 2015]. Fung et al. [2016]
researched the spread of misinformation about disease outbreaks to inform public
health communication strategies.
vary across the source and target domains; if not handled well it can lead to the nega-
tive transfer of knowledge. Our work on cross-domain sentiment classification varies
from Tan et al. [2011] as we carefully model the user behavior based on the related-
ness between the source and target domains, which prevents the ‘negative transfer’.
"I wish the life of K'wo-Fan to be written and given into the care
of the Imperial historiographers, that the memory of a life so
beautiful may be preserved in the national annals. His son will inherit
the title of count, and I give him dispensation from an audience.
"I appoint Ho-Ching, lieutenant-general of Kiang-Su, to be
instructor of the children and grand-children of the deceased. A
token of my munificence will be given to them, that they may know
how my throne remembers and honours a loyal servant.
"Let this edict be respected!"
The homage rendered to heroes, wise men, and philanthropists,
has its origin in the religious principles inculcated by Chinese
philosophers. These philosophers were very numerous in China in
past days, and it is only possible to give an account here of the most
celebrated of them.
Some twenty-three centuries before the Christian era the Chinese
simply worshipped one Supreme Being, first under the vague name
of Thian, or Heaven; later under the more personal title of Ti Shang,
or the Great One.
Gradually, however, this monotheism was succeeded by the
deification of the heavenly bodies, each with a priest of its own,
whose business it was to advise those responsible for the
government of China. These priests, who became in course of time
extremely powerful, won their influence through the study of
astronomy; but as that influence sometimes ran counter to the
wishes of the emperors and bid fair to supersede their power, they
eventually suppressed the entire hierarchy. In Europe this
interference with the spiritual guides of the people would have
aroused a passion of fanaticism, and have resulted in massacres and
religious wars, but nothing of the kind occurred in China, for there
the martyr's palm and crown are never coveted, and religious zeal
never produces the terrible results with which the student of
European history is familiar. Truly, the Celestials are to be
congratulated on the calmness with which they accept what they
consider the inevitable.
The following characteristic epitome of the religious EMPEROR
A WISE
On the death of his mother he had her body transported to the grave of
his father, saying: "Those who were united in life should not be separated
after death." The two were therefore buried together with their heads
towards the north and their feet towards the south. The remains were
protected from wild beasts by being placed in strongly constructed wooden
coffins, made of planks four inches thick and smeared with oil and varnish.
To ensure their preservation as long as possible mounds of earth forming
regular little hills were piled up above them.
During the three years of mourning which succeeded his sad loss,
when, according to a custom still observed, he could do no public work,
Confucius devoted himself to the study of ancient usage in everything
connected with the death of a father or a mother.
"As man," he wrote, "is the A CHINESE TOMB
"The germ has now become a living being, but this living being is not
yet promoted to the dignity of a man; it does not become one until it is
united with that intellectual substance which Heaven bestows on it to
enable it to understand, to compare, and to judge. So long as this being,
thus animated and endowed with intelligence, continues to combine the
two principles necessary to the development, extension, the growth and
the perfection of its form, it will enjoy life; it ceases to live as soon as these
two principles cease to combine. It does not attain to the fulness of life
except by degrees, and by means of expansion; in the same way it is only
finally destroyed by gradual decay. Its destruction is not, however,
destruction properly so called, it is a decomposition into its original
elements; the intellectual substance returns to the heaven whence it came;
the animal breath, or the Khi, becomes united with the aërial fluid, whilst
the earthly and liquid substances become once more earth and water.
"Man, say our ancient sages, is a unique being, in whom are THE NATURE
MAN
OF
be salutary to relate one anecdote illustrating the view the reformer took of
the matter, now that so many despairing souls have lost the aids and
consolations of religious faith in struggling with the difficulties of their life
on earth; when followers of the stoical and heroic Zeno are becoming rarer
and rarer, and so many young men and women resort to the fumes of
charcoal, or to the waters of the nearest river, to put an end to the woes
they have not the courage to face. We must premise, however, that there is
really far more excuse for an Asiatic to take his own life than for a
European, there being nothing unreasonable about it according to the
doctrine of Buddha, whose disciples believe firmly in the transmigration of
souls. They do not, it is true, profess to know whether, if they commit
suicide they will become animals, but they are firmly convinced that they
will continue to live, whereas the atheist has faith in nothingness alone.
In one of his many journeys Confucius and his
disciples met a man who was trying to strangle
himself with a rope. When asked what his motives
were for wishing to commit suicide, he replied that
he had been a bad son, a bad father, and a bad
citizen. The remorse he felt for the terrible
character his self-examination revealed him to be
from all these three points of view, had made his
life odious to him, and he had come out to a lonely
place to put an end to it.
Greatly shocked, Confucius reproved him,
addressing him in the following terms: "However
great the crimes you have committed, the worst of
all of them is yielding to despair. All the others may
FIG. 31.—A DESPERATE be allowed, but that is irremediable. You have, no
MAN. doubt, gone astray from the very first steps you
took upon earth. You should have begun by being
a man of ordinary worth before attempting to
distinguish yourself You cannot attain to being an eminent person until you
have strictly fulfilled the duty imposed by nature on every human creature.
You ought to have begun by being a good son; to love and serve those to
whom you owed your being was the most essential of your obligations; you
neglected to do so, and from that negligence have resulted all your
misfortunes.
SPEECH OF
CONFUCIUS
"Do not, however, suppose that all is lost; take courage again, and try
to become convinced of a truth which all past centuries have proved to be
incontestable. This is the truth I refer to; treasure it up in your mind, and
never lose hold of it: As long as a man has life, there is no reason to
despair of him; he may pass suddenly from the greatest trouble to the
greatest joy, from the greatest misfortune to the greatest felicity. Take
courage once more, return home, and strive to turn to account every
instant, as if you began to-day for the first time to realize the value of life."
Then turning to the younger of his disciples, Confucius said to them:
"What you have heard from the lips of this man is an excellent lesson for