Department of Computer Science and

Engineering

GUIDELINES FOR B. TECH PROJECT REPORT PREPARATION

This document is intended to provide a set of specific and uniform guidelines to the
B. Tech students in the preparation of the eighth semester project report. The content of the
report, which is submitted to the Department of Computer Science and Engineering in partial
fulfillment for the award of the degree of Bachelor of Technology, is very much important.
It is also imperative that the report, to be acceptable by the Department, should essentially
meet a uniform format emphasizing readability, concordance with ethical standards and
University
-wide homogeneity.

1 REPORT LAYOUT
The report has to be organized in the following order.
1. Cover Page
2. Inside Title Page
3. Certificate signed by the Supervisor(s) (in the stipulated format)
4. Declaration signed by the Candidate (in the stipulated format)
5. Acknowledgements
6. Abstract
7. Table of Contents
8. List of Figures
9. List of Tables
10. Abbreviations/ Notations/ Nomenclature (if any)
11. Text of the Report
Chapter 1
Chapter 2
…..
…..
12. References
13. Appendices (if any)

1
 Department of Computer Science and
Engineering

The formats to be followed for various headings are as follows

1. COVER PAGE: See sample sheet 1. The content, relative font size and locations
of various items in the page should match with those given in sample sheet 1. All
copies to be submitted to the Department should be soft bound with white cardboard and
black rexin in spine.

2. INSIDE TITLE PAGE: Same as that of cover page except but printed in
bond paper as given in 2.3.

3. CERTIFICATE: See sample sheet 2. The content, relative font size and locations
of various items in the page should match with those given in sample sheet 2.

4. DECLARATION: See sample sheet 3. The content, relative font size and locations
of various items in the page should match with those given in sample sheet 3.

5. ACKNOWLEDGEMENTS: See sample sheet 4. Should not exceed two pages.

6. ABSTRACT: See sample sheet 5.

7. TABLE OF CONTENTS: See sample sheet 6.
8. LIST OF FIGURES: See sample sheet 7.
9. LIST OF TABLES: See sample sheet 8
10. ABBREVIATIONS/ NOTATIONS/ NOMENCLATURE:
See sample sheet 9.

11. CHAPTERS: The chapters may have Introduction including literatures referred,
Materials, Methods used, Results, Discussions and Conclusions. See sample sheet 10.

12. REFERENCES: To be provided immediately after the last chapter.

See sample sheet 11.

13. APPENDICES: See sample sheet 12.

Each of the items - Inside cover page, Certificate, Acknowledgements, Abstract,
Table of Contents, List of Figures, List of Tables, Abbreviations, Notations,
Nomenclature, each new Chapter, References and each new Appendix should start on
the right side.

2
 Department of Computer Science and
Engineering

2 GENERAL GUIDELINES
2.1. Report Size
Report may contain maximum of about 100 pages including references and
appendices.

2.2. Paper Size

Use A4 size paper (210 mm wide and 297 mm long).

2.3. Paper Quality

White bond paper weighing 85 g/m 2 or more should be used. Essentially the same
quality of paper should be used throughout. Photographs or images with dense colors
may be printed in single side on glossy paper.

2.4. Margins
A margin of 35 mm is to be provided on left and right sides, whereas top and
bottom margins should be 30 mm. No print matter should appear in the margin except
the page numbers. All page numbers should be centered inside the bottom margin,
20mm from the bottom edge of the paper.

2.5. Font
Times New Roman (TNR) 12 point font has to be used throughout the running
text. The captions for tables and figures should have font size of 11 and foot notes
should be set at font size 10. Font sizes for various levels of headings are given in
section 2.7.

2.6. Line Spacing

The line spacing in the main text should be 1.5. Single line spacing
should be given for quotations, abstract, figure captions, table captions, figure legends,
footnotes, and references. The equations, tables, figures, and quotations should be
set off from the main text both before and after with spacing of 1.5. Two consecutive
paragraphs should be separated by triple line spacing.

2.7. Headings
Following format has to be followed in heading of chapters and sections.

3
 Department of Computer Science and
CHAPTER 3
TITLE PAGE-CENTERED TNR 17-POINT BOLD ALL CAPS
Section Heading
Left aligned with number, TNR 17 points, bold and leading caps

Second level section heading

Left aligned with number, TNR 14 points, bold and sentence case.

Third level section heading

Left aligned with number, TNR 12 points, bold and sentence case.
Fourth-level section heading
Numbered subsections beyond third level are not recommended. However,
Fourth-level subsection headingsmay be included without numbering, TNR 12-point font, left aligned and italicized.

Running text should be set in 12-point TNR and fully justified. First line of
paragraph should have indentation of 15 mm.

2.8. Table / Figure/equation Format

Tables, figures and equations shall be numbered chapter-wise. For example,
second figure in Chapter 3 will be numbered Figure 3.2. The figure can be cited in the
text as Fig. 3.2 or Figure 3.2, however consistent citation format should be followed
throughout the thesis. Tables shall be numbered similarly (Table 2 in Chapter 3 will be
numbered Table 3.2) and shall be cited in the text as Table 3.2. Figure caption shall be
located below the figure. Table number and caption shall be located above the table.
Equations are aligned to the center of the page with equation number in the text has to be
given at the end of the line within brackets as given below.

4
 Department of Computer Science and

2.9. Citing References

2.9.1. One author
Monika (2007) developed this method of ……… Subsequently other
researchers have adopted this technique (Ramakrishna, 2009; Bhaskar, 2010).

2.9.2. Two authors

Monika and Ram (2008) developed the model of.............Subsequently other
researchers adopted this technique (Ramakrishna, 2009; Rao and Ram,
2011)

2.9.3. Multiple authors & more than one publication in a year

“Ram et al. (2005a) has designed the model …..” when given in
sentence.
“Model AAB could regulate the control unit more efficiently (Ram et
al., 2005b) ….” while given in brackets

2.9.4. Citing multiple references

When many authors are cited in sentence it is given as “..............Similar work was
also proposed by Singh and Robin (2008); Ram et al. (2009); Prakash (2011)....”

“Similar work was demonstrated for varied data set by many

researchers (Singh and Robin, 2008; Ram et al., 2009; Prakash, 2011) ”

2.10. Listing of the References

References are to be listed after last chapter. They are to be listed in
alphabetical order and numbered. Within a reference the line spacing should be single.
Each reference should be separated by one blank line. The reference number should be
left aligned. The text of the reference should have an indentation of 10 mm. The
reference format to be followed for journal articles, text books, conference proceedings
etc. are given below.

5
 Department of Computer Science and

2.10.1. Journals
1. Prakas, K. (2011). Feedback and optimal sensitivity: Model reference
transformations, multiplicative seminorms, and approximate inverses. IEEE
Transactions on Automatic Control, 26(2): 301–320.

2. Ram, R., Krishna, S. and Peter, K. (2005a). Risk sensitive estimation and a
differential game. IEEE Transactions on Automatic Control, 39(9): 1914–
1918.

3. Ram, R., Krishna, S. and Peter, K. (2005b). Differential rectification using control
points. IEEE Transactions on Geoscience and Remote sensing, 55:
914 – 918.

4. Ram, R., Krishnamurthy, P., Prasad, N. and Peter, K. (2009). Risk sensitive
estimation model II. IEEE Transactions on Automatic Control, 43(15):
355 - 363.

2.10.2. Text books

1. Myers, D. G. (2007). Psychology (1st Canadian ed.). Worth: New York.

2. Robin, R. (2008). Robust Statistics. Wiley-Interscience: New York.

2.10.3. Conference proceedings

1. Payne, D.B. and Gunhold, H.G. (1986). Digital sundials and broadband
technology, In Proc. IOOC-ECOC, 1986, pp. 557-998.

2. Singh, K. and Robin, R. (2008). A linear- quadratic game approach to estimation

and smoothing. In American Control Conference, New York. June 20 – 25, 2008,
pp. 2818–2822.

2.10.4. Reports
1. Milton, M and Robert, L. (2004). Atmospheric carbon emission through genetic
algorithm, Environment and Technical Report No.3., Indian Meteorological
Department., New Delhi.

2.10.5. Online journals with a DOI (Digital Object Identifier)

1. Krebs, D.L. and Denton, K. (2006). Explanatory limitations of cognitive-
developmental approaches to morality. Psychological Review, 113(3): 672-
675. doi: 10.1037/0033-295X.113.3.672

2.10.6. Online journals without a DOI

1. Vicki, G.T., Thomae, M., Cullen, A. and Fernandez, H. (2007). Modeling the
hydrological impact on Tropical Forests. Forest Ecology, 13(10): 122-132.
Retrieved from http://www.uiowa.edu/~grpproc/crisp/crisp.html
6
 Department of Computer Science and

2.10.7. Online abstracts

1. Perilloux, C. and Buss, D.M. (2008). Human relationships: Costs experienced and
coping strategies deployed. Evolutionary Psychology, 6(1): 164-181. Abstract
retrieved from http://www.epjournal.net

2.10.8. Online books

1. Perfect, T.J. and Schwartz, B. L. (Eds.) (2002). Applied metacognition.
Retrieved from http://www.questia.com/read/107598848 (--If DOI is
available, use the DOI instead of a URL)

2.10.9. Chapters from a book

1. Krebs, D.L. and Denton, K. (1997). Social illusions and self deception: The evolution
of biases in person perception. In J. A. Simpson & D. T. Kenrick (Eds.),
Evolutionary social psychology (pp.21-48). Hillsdale, NJ: Erlbaum.

2.10.10. Books in print form

1. Snyder, C.R., Higgins, R.L. and Stucky, R.J. (Eds.). (1983). Excuses:
Masquerades in search of grace. New York, NY: John Wiley & Sons.

2.10.11. Dissertations and Thesis

1. Mack, S. (2000). “Desperate Optimism” M.S. Thesis, University of
Calgary, Canada.

2.11. Page Numbering

Page numbers for the prefacing materials (Inside title page, dedication, certificate, declaration,
acknowledgements, abstract, table of contents, etc.) of the thesis shall be in small Roman numerals
and should be centered at the bottom of the pages. The numbering of the prefacing material
starts from the Inside Title Page. However, the number is not printed on the Inside Title Page. Each
new item of the prefacing materials listed above should start on a fresh paper on right page. If
the content of the prefacing material exceeds one page, it has to be printed on only one side of
the paper by starting from the right side page. For example, if the item “Table of Contents‟
extends for 5 pages, it should be printed in fresh paper on right side page with second page of the
“Table of Contents‟ on the next right side of the paper and then continued. The page numbers of the
prefacing material will be printed in small Roman numerals continuously counting blank pages
also. However, the numbers are not printed on the blank pages.

7
 Department of Computer Science and

The body of the thesis starting from Chapter 1 should be paginated in Arabic numerals and
should be centered at the bottom of the pages. The pagination should start with the
first page of Chapter 1 and should continue throughout rest of the thesis.

2.12. Printing
Printing of all material in general should be single –sided in black ink with
exceptions as indicated in sections 2.3 and 2.11.

2.13. Binding
Thesis copies to be submitted for evaluation are to be soft bounded. The cover
page should be printed on glossy white card of 300 g/m 2 or above. The number of copies
to be submitted = number of students in a batch + 2 ( E.g. 4 students in a batch + 2 = 6
copies )

2.14. Electronic Copy

An electronic version of the report should be submitted to the form shared by
the project coordinator the file name should contain the BATCH Name in the form of
pdf.

8
 Sample sheet 1

TITLE OF THE PROJECT REPORT

A project Report submitted

in partial fulfillment for the award of the Degree of

Bachelor of Technology
in
Computer Science and Engineering
by

D.MAHENDRA REDDY (U17CS182)

K. SOUJANYA (U17CS224)
C.VASANTH KUMAR (U17CS162)
AJU (U17CS198)

Under the guidance of

Ms. S. POTHUMANI., M.E., (Ph.D)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SCHOOL OF COMPUTING

BHARATH INSTITUTE OF HIGHER EDUCATION AND RESEARCH

(Deemed to be University Estd u/s 3 of UGC Act, 1956)

CHENNAI 600 073, TAMILNADU, INDIA

April , 2022
 Sample sheet 2

CERTIFICATE

This is to certify that the project report entitled <Title of the project>
submitted by <Name of the candidate(s)> to the Department of Computer
Science and Engineering, Bharath Institute of Higher Education and
Research, in partial fulfillment for the award of the degree of B. Tech in
(Computer Science and Engineering) is a bona fide record of project work
carried out by them under my supervision. The contents of this report, in full
or in parts, have not been submitted to any other Institution or University for
the award of any other degree.

<Signature of Supervisor>
<Name>
Department of Computer Science & Engineering,
School of Computing,
Bharath Institute of Higher Education and Research
April, 2022

<Signature of Head of the Department>

Dr. B. Persis Urbana Ivy

Professor & Head
Department of Computer Science & Engineering,
School of Computing,
Bharath Institute of Higher Education and Research,

April, 2022
 Sample sheet 3

DECLARATION

We declare that this project report titled <Title of the report> submitted in
partial fulfillment of the degree of B. Tech in (Computer Science and
Engineering) is a record of original work carried out by us under the
supervision of <Name of the Supervisor>, and has not formed the basis for
the award of any other degree or diploma, in this or any other Institution
or University. In keeping with the ethical practice in reporting scientific
information, due acknowledgements have been made wherever the findings
of others have been cited.

<Signature>
<Name of the candidate>
< (Student Reg. Number)>

<Signature>
<Name of the candidate>
< (Student Reg. Number)>

<Signature>
<Name of the candidate>
< (Student Reg. Number)>

<Signature>
<Name of the candidate>
< (Student Reg. Number)>

Chennai
<Date>
 Sample sheet 4

ACKNOWLEDGMENTS
First, we wish to thank the almighty who gave us good health and success throughout our
project work.

We express our deepest gratitude to our beloved President Dr. J. Sundeep Aanand, and
Managing Director Dr.E. Swetha Sundeep Aanand for providing us the necessary facilities for
the completion of our project.

We take great pleasure in expressing sincere thanks to Vice Chancellor (I/C) Dr. K.
Vijaya Baskar Raju, Pro Vice Chancellor (Academic) Dr. M. Sundararajan, Registrar Dr. S.
Bhuminathan and Additional Registrar Dr. R. Hari Prakash for backing us in this project.
We thank our Dean Engineering Dr. J. Hameed Hussain for providing sufficient facilities for
the completion of this project.

We express our immense gratitude to our Academic Coordinator Mr. G. Krishna

Chaitanya for his eternal support in completing this project.

We thank our Dean, School of Computing Dr. S. Neduncheliyan for his encouragement
and the valuable guidance.

We record indebtedness to our Head, Department of Computer Science and

Engineering Dr. B. Persis Urbana Ivy for immense care and encouragement towards us
throughout the course of this project.

We also take this opportunity to express a deep sense of gratitude to our Supervisor Ms.
S. Pothumani for her cordial support, valuable information and guidance, she helped us in
completing this project through various stages.

We thank our department faculty, supporting staff and friends for their help and
guidance to complete this project.

D.MAHENDRA REDDY (U17CS182)

K.SOUJANYA (U17CS224)
C.VASANTH KUMAR (U17CS162)
K.YUVARAJU (U17CS198)
 Sample

ABSTRACT

Abstract Agriculture is first and foremost factor which is important for survival. Machine learning
(ML) could be a crucial perspective for acquiring real-world and operative solution for crop yield
issue. Considering the present system including manual counting, climate smart pest management
and satellite imagery, the result obtained arent really accurate. This paper focuses mainly on
predicting the yield of the crop by applying various machine learning techniques. The classifier
models used here include Logistic Regression, Naïve Bayes and Random Forest, out of which the
Random Forest provides maximum accuracy. The prediction made by machine learning algorithms
will help the farmers to come to a decision which crop to grow to induce the most yield by
considering factors like temperature, rainfall, area, etc. This bridges the gap between technology and
agriculture sector.
KeywordsCrop_yield_prediction; logistic_regression; naïve bayes; random forest; weather_api

i
 Sample

TABLE OF CONTENTS

DESCRIPTION PAGE NUMBER

CERTIFICATE iii
DECLARATION v
ACKNOWLEDGEMENTS vii
ABSTRACT ix
LIST OF FIGURES xiii
LIST OF TABLES xv
ABBREVIATIONS/ NOTATIONS/ NOMENCLATURE xvii
1. TITLE OF CHAPTER 1 1
1.1 Section heading name 1
1.2 Section heading name 1
1.2.1 Second level section heading 3
1.3 Section heading name 4
1.4 Section heading name 5
1.4.1 Second level section heading 8
1.4.2 Second level section heading 11
1.4.2.1 Third level section heading 20
2. TITLE OF CHAPTER 2 23
2.1 Section heading name 23
2.2 Section heading name 24
2.2.1 Second level section heading 25
2.3 Section heading name 26
2.4 Section heading name 28
2.4.1 Second level section heading 30
2.4.2 Second level section heading 35

3. TITLE OF CHAPTER 3 41
3.1 Section heading name 41

x
 Sample
3.2 Section heading name 44
3.3 Section heading name 50
3.4 Section heading name 52
3.4.1 Second level section heading 59
3.4.2 Second level section heading 65
3.4.2.1 Third level section heading 70
4. TITLE OF CHAPTER 4 75
4.1 Section heading name 75
4.2 Section heading name 79
4.2.1 Second level section heading 89
4.3 Section heading name 93
4.4 Section heading name 101
4.4.1 Second level section heading 126
4.4.2 Second level section heading 150
4.4.2.1 Third level section heading 190
5. TITLE OF CHAPTER 5 207
5.1 Section heading name 211
5.2 Section heading name 290
5.3 Section heading name 311
5.3.1 Second level section heading 329
5.3.2 Second level section heading 330
5.3.2.1 Third level section heading 340
REFERENCES 349
Appendix 1 Title of the appendix 1 361
Appendix 2 Title of the appendix 2 369

Sample sheet 7

x
 LIST OF FIGURES

FIGURE TITLE PAGE NUMBER

1.1. Schematic diagram showing the 5

proposed mechanism

1.2. Next figure alpha 9

2.1. Next figure beta 35

2.2. Next figure gamma 39

3.1. Next figure alpha, alpha 46

4.1. Plot of concentration vs time 120

5.1. Next figure beta, beta 310

x
 Sample sheet

LIST OF TABLES

TABLE TITLE PAGE NUMBER

1.1. Percentage composition of 11

fuel components

1.2. Name of Table delta 12

2.1. Name of Table gradient 35

4.2. Name of Table alpha 100

5.1. Kinetic parameters calculated for the system 310

x
 Sample sheet

ABBREVIATIONS/ NOTATIONS/ NOMENCLATURE

Utmost care should be taken by the project student while using technical abbreviations,
notations and nomenclature.

The abbreviations should be listed in alphabetical order as shown below.

AFM Atomic Force Microscopy

BBB Blood Brain Barrier

CNT Carbon Nanotube

The meaning of special symbols and notations used in the report should be explained.

|x| - absolute value of x

μ - mean

logn(x) - logarithm (x) to the base n

Chemical nomenclature

NH4+ - ammonium

CH4 - methane

OH- - hydroxide

SO42- - sulphate

Biological nomenclature

Soneratia apetalla - saline tolerant species

Oryza sativa - common rice

x
 Sample sheet

CHAPTER 1
INTRODUCTION

INTRODUCTION

1. Agriculture, since its invention and inception, be the prime and pre-eminent activity of every culture and
civilization throughout the history of mankind. It is not only an enormous aspect of the growing economy, but
its essential for us to survive. Its also a crucial sector for Indian economy and also human future. It also
contributes an outsized portion of employment. Because the time passes the requirement for production has
been increased exponentially. So as to produce in mass quantity people are using technology in an exceedingly
wrong way. New sorts of hybrid varieties are produced day by day. However, these varieties dont provide the
essential contents as naturally produced crop. These unnatural techniques spoil the soil. It all ends up in further
environmental harm. Most of these unnatural techniques are wont to avoid losses.
2. But when the producers of the crops know the accurate information on the crop yield it minimizes the loss.
Machine learning, a fast-growing approach thats spreading out and helping every sector in making viable
decisions to create the foremost of its applications. Most devices nowadays are facilitated by models being
analyzed before deployment. The main concept is to increase the throughput of the agriculture sector with the
Machine Learning models. Another factor that also affects the prediction is the amount of knowledge thats
being given within the training period, as the number of parameters was higher comparatively. The core
emphasis would be on precision agriculture, where quality is ensured over undesirable environmental factors.
So as to perform accurate prediction and stand on the inconsistent trends in
3. temperature and rainfall various machine learning classifiers like Logistic Regression, Naïve Bayes, Random
Forest etc. are applied to urge a pattern. By applying the above machine learning classifiers, we came into a
conclusion that Random Forest algorithm provides the foremost accurate value. System predicts crop prediction
from the gathering of past data. Using past information on weather, temperature and a number of other factors
the information is given. The Application which we developed, runs the algorithm and shows the list of crops
suitable for entered data with predicted yield value.

LITERATURE SURVEY
4. Aruvansh Nigam, Saksham Garg, Archit Agrawal[1] conducted experiments on Indian government dataset and
its been established that Random Forest machine learning algorithm gives the best yield prediction accuracy.
Sequential model thats Simple Recurrent Neural Network performs better on rainfall prediction while LSTM is
good for temperature prediction. The paper puts factors like rainfall, temperature, season, area etc. together for
yield prediction. Results reveals that Random Forest is the best classier when all parameters are combined.
5. Leo Brieman [2] , is specializing in the accuracy and strength & correlation of random forest algorithm.
Random forest algorithm creates decision trees on different data samples and then predict the data from each
subset and then by voting gives better the answer for the system. Random Forest used the bagging method to
trained the data. To boost the accuracy, the randomness injected has to minimize the correlation while
maintaining strength.
6. Balamurugan [3], have implemented crop yield prediction by using only the random forest classifier. Various
features like rainfall, temperature and season were taken into account to predict the crop yield. Other machine
learning algorithms were not applied to the datasets. With the absence of other algorithms, comparison and
quantification were missing thus unable to provide the apt algorithm.
7. Mishra [4], has theoretically described various machine learning techniques that can be applied in various
1
 Sample sheet
forecasting areas. However, their work fails to implement any algorithms and thus cannot provide a clear insight
into the practicality of the proposed work.
8. Dr. Y. Jeevan Nagendra Kumar [5], have concluded Machine Learning algorithms can predict a target/outcome
by using Supervised Learning. This paper focuses on supervised learning techniques for crop yield prediction.
To get the
9. specified outputs it needs to generate an appropriate function by set of some variables which can map the input
variable to the aim output. The paper conveys that the predictions can be done by Random Forest ML algorithm
which attain the crop prediction with best accurate value by considering least number of models.

METHODOLOGY
a. Data Pre-Processing
b. Data Preprocessing is a method that is used to convert the raw data into a clean data set. The data are gathered
from different sources, it is collected in raw format which is not feasible for the analysis. By applying different
techniques like replacing missing values and null values, we can transform data into an understandable format.
The final step on data preprocessing is the splitting of training and testing data. The data usually tend to be split
unequally because training the model usually requires as much data- points as possible. The training dataset is
the initial dataset used to train ML algorithms to learn and produce right predictions (Here 80% of dataset is
taken as training dataset). Fig.1. shows the few rows of the preprocessed data.
c. Factors affecting Crop Yield and Production
d. There are a lot of factors that affects the yield of any crop and its production. These are basically the features
that help in predicting the production of any crop over the year. In this paper we include factors like
Temperature, Rainfall, Area, Humidity and Windspeed (Fig.1 shows the attributes for the crop name prediction
and its yield calculation).
e. Fig. 1. Preprocessed data
f. Comparison and Selection of Machine Learning Algorithm
g. Before deciding on an algorithm to use, first we need to evaluate and compare, then choose the best one that fits
this specific dataset. Machine Learning is the best technique which gives a better practical solution to crop yield
problem. There are a lot of machine learning algorithms used for predicting the crop yield. In this paper we
include the following machine learning algorithms for selection and accuracy comparison :
 .Logistic Regression:- Logistic regression is a supervised learning classification algorithm used to predict the
probability of target variable. The nature of target or dependent variable is dichotomous, which means there
would be only two possible classes. When logistic regression algorithm applied on our dataset it provides an
accuracy of 87.8%.
 Naive Bayes:- Naive Bayes classifier assumes that the presence of a particular feature in a class is unrelated to
the presence of any other feature. Naive Bayes model is easy to build and particularly useful for very large data
sets. Along with simplicity,
 Naive Bayes is known to outperform even highly sophisticated classification methods. It provides an accuracy
of 91.50%.
 Random Forest:- Random Forest has the ability to analyze crop growth related to the current climatic conditions
and biophysical change. Random forest algorithm creates decision trees on different data samples and then
predict the data from each subset and then by voting gives better solution for the system. Random Forest uses
the bagging method to train the data which increases the accuracy of the result. For our data, RF provides an
accuracy of 92.81%.
 It is clear that among all the three algorithms, Random forest gives the better accuracy as compared to other
algorithms.
h. Random Forest Model for Crop Prediction
i. Fig. 2. Flowchart for Random Forest Model
j. Random forests are the aggregation of tree predictors in such a way that each tree depends on the values of a
random subset sampled independently and with the same distribution for all trees in the forest. Random Forest
used the bagging method to trained the data which increases the accuracy of the result. For getting high
accuracy we used the Random Forest algorithm which gives accuracy which predicate by model and actual
1
 Sample sheet
outcome of predication in the dataset. The predicted accuracy of the model is analyzed 91.34%. Fig.2 shows the
flowchart of random forest model for crop yield prediction.
k. System Architecture
l. System architecture represented in the Fig.3 mainly consists of weather API where we fetch the data such as
temperature, humidity, rainfall etc. The data fetched from the API are sent to the server module. The data gets
stored on to the database on the server. Using the mobile application, the user can provide details like location,
area, etc. The user can create an account on the mobile app by one-time registration
m. and all these entered data are sent to server. The trained Random forest model deployed on the server uses all
the fetched and input data for crop yield prediction, finds the yield of predicted crop with its name in the
particular area.
n. Fig. 3. System Architecture
o. Proposed System
p. Our proposed system system is a mobile application which predicts name of the crop as well as calculate its
corresponding yield. Name of the crop is determined by several features like temperature, humidity, wind-
speed, rainfall etc. and yield is determined by the area and production. In this paper, Random Forest classifier is
used for prediction. It will attain the crop prediction with best accurate values.
q. System Analysis
 Python 3.8.5(Jupyter Notebook):Python is the coding language used as the platform for machine learning
analysis. Jupyter Notebooks illustrates the analysis process and gives out the needed result.
 Weather_API (Open Weather Map): Weather API is an application programming interface used to access the
current weather details of a location. The generated API key illustrates current weather forecast needed for crop
prediction.
 Android Studio (Version 3.4.1): Android Studio is the official integrated development environment (IDE) for
Android application development. This paper uses java as the framework for frontend designing. USB
debugging method is used for the connection of IDE and app.
 Python Flask Framework (Version 2.0.1): Flask is a micro framework in python. Flask is based on WSGI(Web
Server Gateway Interface) toolkit and Jinja2 template engine. In this paper flask is used as the back-end
framework for building the application. It is the collection of modules and libraries that helps the developer to
write applications without writing the low-level codes such as protocols, thread management, etc.
 Heroku: Heroku is the container-based cloud platform that allows developers to build, run & operate
applications exclusively in the cloud. In this paper Heroku is used for server part. Once created an account in
the Heroku we can connect it with the GitHub repository and then deploy.

RESULTS AND DISCUSSIONS

This paper reinforces the crop production with the aid of machine learning techniques. The technique which
results in high accuracy predicted the right crop with its yield. The machine learning algorithms are implemented on
Python 3.8.5(Jupyter Notebook) having input libraries such as Scikit- Learn, Numpy, Keras, Pandas. Developed
Android application queried the results of machine learning analysis. Flutter based Android app portrayed crop name
and its corresponding yield.
r. Datasets Used
s. The datasets have been obtained from different official Government websites:
 data.gov.in-Details regarding area, production, crop name[8].
 indianwaterportal.org -Depicts rainfall details[9].
 power.larc.nasa.in Temperature, humidity, wind speed details[10].
 Combined dataset has 4261 instances. It includes features like crop name, area, production, temperature,
rainfall, humidity and wind speed of fourteen districts in Kerala. The data pre- processing phase resulted in
needed accurate dataset. Fig. 4. shows a heat map used to portray the individual attributes contained in.
 Fig. 4. Heat Map

1
 Sample sheet
t. Classifiers Used
u. Machine learning classifiers used for accuracy comparison and prediction were Logistic Regression, Random
Forest and Naïve Bayes. These three classifiers were trained on the dataset
v. Fig. 5.Comparison Plot
w. and a comparison graph was plotted to showcase the performance of the models. Fig.5 showcase the
performance of the models. Of the three classifiers used, Random Forest resulted in high accuracy.
x. Weather_ API Used
y. Weather _ API usage provided current weather data access for the required location. For retrieving the weather
data used API
z. Fig. 6. API Output
aa. was OpenWeatherMap. By entering the district name, needed metrological factors such as near surface elements
which include temperature, wind speed, humidity, precipitation were accessed by using generated API key.
Fig.6. depicts current weather description for entered location.
bb. After the training of dataset, API data was given as input to illustrate the crop name with its yield.
cc. Crop Name Prediction
dd. Random Forest Classifier having the highest accuracy was used as the midway to predict the crop that can be
grown on a selected district at the respective time.
ALGORITHM ACCURACY
RANDOM FOREST 92.81407991690006
NAÃVE BAYES 91.49621790098573
LOGISTIC REGRESSION 87.82982929223341
ee. Table I : Accuracy Table
ff. Abundantly growing crops in Kerala were chosen and their name was predicted and yield was calculated on the
basis of area, production, temperature, humidity, rainfall and wind speed. The preprocessed dataset was trained
using Random Forest classifier. Chosen districts instant weather data accessed from API was used for
prediction. Trained model resulted in right crop prediction for the selected district.
gg. Crop Yield Calculation
hh. The crop which was predicted by the Random Forest Classifier was mapped to the production of predicted crop.
Then the area entered by the user was divide from the production to get crop yield[1].
ii. Yield= Production/Area
jj. Crop name predictedwith their respective yield helps farmers to decide correct time to grow the right crop to
yield maximum result.
kk. Android Application
10. An Android app has been developed to query the results of machine learning analysis. The app is compatible
with Android OS version 7. The pages were written in Java language. The app has a simple, easy-to-use
interface requiring only few taps to retrieve desired results. Just only giving the location and area of the field the
Android app gives the name of right crop to grown there.
11. By accessing the user entered details, app will queries the machine learning analysis. Using the location, API
will give out details of weather data. The retrieved weather data get acquired by machine learning classifier to
predict the crop and calculate the yield. The output is then fetched by the server to portray the result in
application.
12. The main activities in the application were account creation, detail_entry and results_fetch. The
account_creation helps the user to actively interact with application interface. The user fill the field in home
page to move onto the results activity. The retrieved data passed to machine learning model and crop name is
predicted with calculated yield value.
13. Fig. 7. Home Screen

1
 Sample sheet

CONCLUSION
14. This paper focuses on the prediction of crop and calculation of its yield with the help of machine learning
techniques. Several machine learning methodologies used for the calculation of accuracy. Random Forest
classifier was used for the crop prediction for chosen district. Implemented a system to crop prediction from the
collection of past data. The proposed technique helps farmers in decision making of which crop to cultivate in
the field. This work is employed to search out the gain knowledge about the crop that can be deployed to make
an efficient and useful harvesting. The accurate prediction of different specified crops across different districts
will help farmers of Kerala. This improves our Indian economy by maximizing the yield rate of crop
production.

FUTURE SCOPE
In coming years, can try applying data independent system. That is whatever be the format our system should
work with same accuracy. Integrating soil details to the system is an advantage, as for the selection of crops knowledge
on soil is also a parameter. Proper irrigation is also a needed feature crop cultivation. In reference to rainfall can depict
whether extra water availability is needed or not. This research work can be enhanced to higher level by availing it to
whole India.

Table 1.1 Title of the table (Times New Roman 11)

Aa Bb C D

a
A is admonishment coefficient of total population (Times New Roman 10)

b
B is Bombardment coefficient of the mean population (Times New Roman 10)

1.1.1 Motivation of the study

Alarming rate of climate change, sea level rise and other natural disasters are to
be managed efficiently. Assessment and management of green house gases thus
become very much essential..

1
Adapted from Monika and Ram, 2008 (Times New Roman 10)

1
 Sample sheet

The satellite image as given in Figure 1.1 shows the area from where
samples are collected.

Figure 1.1 Title of the figure (Times New Roman 11)

1
 Sample sheet

REFERENCES
1. Attanas, D.B. and Monica, H.G. (2012). Effects of green house gases, In Proc.
IOOC-ECOC, pp. 557-998.

2. Gurudeep, P.R. and Mahin, P. (2009). Risk sensitive estimation model II.
IEEE Transactions on Automatic Control, 43 (15): 355 - 363.

3. Prakas, K. (2011). Feedback and optimal sensitivity: Model reference

transformations, multiplicative seminorms, and approximate inverses. IEEE
Transactions on Automatic Control, 26(2): 301–320

4. Ram, R., Krishna, S. and Peter, K. (2005a). Risk sensitive estimation and a
differential game. IEEE Transactions on Automatic Control, 39(9): 1914–
1918.

5. Ram, R., Krishna, S and Peter, K. (2005b). Differential rectification using

control points. IEEE Transactions on Geoscience and Remote sensing, 55: 914
– 918.

6. Singh, K. and Robin, R. (2008). A linear- quadratic game approach to estimation

and smoothing. In American Control Conference, New York. June 20 – 25,
2008, pp. 2818–2822.

3
 Sample sheet 13

APPENDIX 1