AGRI-PREDICT AI-Driven Crop Yield and Price Forecasting

A CAPSTONE PROJECT REPORT

Submitted in partial fulfillment of the

requirement for the award of the
Degree of

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE & ENGINEERING

by

Parepalli Dhyanesh (21BCE9397)

Neelam Suvarna Jyothi (21BCE9552)
Kondapalli Vasavi(21BCE9474)
Shaik Rokiya (21BCE9546)

Under the Guidance of

Prof. Srinivasa Rao Pokuri

SCHOOL OF ELECTRONICS ENGINEERING

VIT-AP UNIVERSITY
AMARAVATI- 522237

DECEMBER 2024
1
 ACKNOWLEDGEMENTS

We would like to express our heartfelt gratitude to our professor Dr.Srinivasrao Pokuri sir, for
his valuable guidance, encouragement, and support throughout the development of this project.
His insights and expertise were instrumental in shaping the project's direction and ensuring its
successful completion. We are also deeply thankful to VIT-AP University for providing a
collaborative environment that fostered innovation and learning. Finally, we extend our sincere
appreciation to our fellow team members for their dedication, hard work, and seamless
collaboration, which made this project possible.

ABSTRACT

The AgriPredict-AI driven crop yield and price forecasting System is a web-based
application designed to assist farmers in making informed decisions about crop selection and yield
estimation. Using modern web development technologies like HTML, CSS, and JavaScript for the
frontend and Flask for the backend, the system provides an intuitive interface for users to input
details such as land area, crop type, fertilizers used, and other relevant factors. A trained machine
learning model processes this data to predict crop yields and estimate potential income. This
software-based solution eliminates the need for additional hardware, making it cost-effective and
accessible to users with an internet connection. By leveraging reliable datasets from trusted sources
such as the FAO and Indian Government portals, the system ensures accurate and actionable
insights, empowering farmers to optimize their agricultural practices and maximize productivity.

2
 TABLE OF CONTENTS
Sl.No. Chapter Title Page Number
1. Acknowledgement 2
2. Abstract 2
3. List of Figures and Table 4
4. 1 Introduction 6
1.1 Objectives 7
1.2 Background and Literature Survey 7
1.3 8
Organization of the Report
5. 2 Chapter Title (Work) 9
2.1 Proposed System 9
2.2 Working Methodology 10
2.3 Standards 10
2.4 System Details 11
2.4.1 Software 11
2.4.2 Hardware 14
6. 3 Cost Analysis 15
3.1 List of components and their cost 15
7. 4 Results and Discussion 16
4.1 Tentative Methodologies 16
4.2 Data Set Information 16
4.3 Processing Data Sets 17
4.4 Exploring Data sets 17
4.4.1 Distribution of data 17
4.4.2 Statistical Description 18
4.5 Visualizing the data to find Insights 18
4.5.1 Correlation matrix 18
4.6 Building pipelines for data preprocessing 21
4.6.1 Numerical Pipeline 21
4.6.2 Categorical Pipeline 22
4.7 Selecting and Training the machine 23
learning model
4.8 Evaluation Metrics 23
4.8.1 PERFORMANCE OF MODELS ON 24
TRAINING SET
EVALUATION OF MODELS ON THE TEST
4.8.2 25
SET
Combining yield with prices
4.9 26
8. 5 Conclusion & Future Works 27
3
9. 6 Appendix 28
10. 7 References 44

List of Tables

Table No. Title Page no.

1 Statistical 18
Representation of data

2. Train R2 Score on 24
Training set

3 Test R2 Score on Test 25

set

4
 List of Figures

Figure Number Title Page no.

1 Agricultural price 6
fluctuations in India

2 System Block Diagram 9

3 Home Page 11
4 Login page 12
5 Sign up page 12
6 Input page 13
7 Guide Page 13
8 Help Page 13
9 Output Page
10 Distribution of Data 17
11 Heat Map 19
12 Scatter Plot 19
13 Bar Graphs 20
14 One Hot Encoding 22

5
 CHAPTER 1
INTRODUCTION

Agriculture remains the cornerstone of the Indian economy, providing livelihoods to millions and
contributing significantly to the nation’s GDP. India’s agricultural landscape includes a diverse
array of crops, such as cereals, pulses, fruits, vegetables, coffee, tea, oilseeds, spices, and non-food
crops like rubber, jute, and cotton. The sector plays a pivotal role in meeting the growing food
demands of an ever-increasing population, yet farmers face persistent challenges, particularly in
price volatility. Fluctuating crop prices hinder farmers’ ability to plan production patterns, causing
economic instability, especially for perishable crops like tomatoes, which have a short shelf life and
are highly sensitive to market price changes. Figure 1 in this report provides a detailed depiction of
crop price fluctuations in India from 2000 to 2024, highlighting the instability in prices over the
years and emphasizing the urgent need for improved agricultural policies and systems to stabilize
market conditions and support farmers.

Figure 1 Agricultural price fluctuations in India

6
1.1 Objectives
The following are the objectives of this project:
• To assist farmers in selecting the most profitable crop to cultivate based on prior crop yield,
climatic conditions, nutrients supplied, and fertilizers used.
• To utilize machine learning methods to predict the best crops for the upcoming season,
ensuring accurate and data-driven decision-making.
• To analyze and interpret weather conditions such as rainfall and temperature, along with
pesticide usage, to forecast crop yields effectively.
• To provide farmers with a tool that calculates and compares the expected yield for various
crops using input attributes, enabling them to identify the highest-yielding crop.
• To empower farmers with reliable information for agricultural risk management, thereby
improving productivity and optimizing resource utilization.

1.2 Background and Literature Survey

1. Machine learning approach for forecasting crop yield based on parameters of climate.
The paper provided in International Conference on Computer Communication and
Informatics (ICCCI). The papers is based on parameters like rainfall, maximum and
minimum temperature, potential evapotranspiration, cloud cover, wet day frequency. C4.5
algorithm is used to produce the most influencing climatic parameter on the crop yields of
selected crops in selected districts of Madhya Pradesh. The paper is implemented using
Decision Tree.
2. Crop production Ensemble Machine Learning model for prediction. International
Journal of Computer Science and Software Engineering (IJCSSE). The parameters
considered in this paper are average temperature, cloud cover, vapor pressure, wet day
frequency, evapotranspiration. In this paper, AdaNaive and AdaSVM are the proposed
ensemble model used to project the crop production over a time period. Implementation
done using AdaSVM and AdaNaive. AdaBoost increases efficiency of SVM and Naive
Bayes algorithm.
3. Random Forests for Global and Regional Crop Yield Predictions, Institute on the
Environment, University of Minnesota, United States of America. The paper considered
parameters like climate, soil, photoperiods, water supply and fertilization data. The
generated outputs show that Random Forests is an effective and different machine-
7
 learning method for crop yield predictions at regional and global scales for its high
accuracy. The Paper is Implemented using k-nearest neighbors, Support Vector
Regression (SVR), Random Forests.
4. Applications of Machine Learning Techniques in Agricultural Crop Production. Indian
Journal of Science and Technology, Vol 9(38). From GPS based color images, model is
provided as an intensified indistinct cluster analysis for classifying plants, soil and residue
regions of interest. The paper includes various parameters which can help the crop yield
for better enhancement and ratio of the yield can be increased during cultivation. The
paper is implemented using Artificial Neural Networks, Time series analysis and Decision
Trees.
5. Predicting yield of the crop using machine learning algorithm. International Journal of
Engineering Science Research Technology. The paper focuses on predicting the yield of
the crop based on the existing data. Real data of Tamil Nadu were used for building the
models and the models were tested with samples. The factors considered are rainfall,
Kharif and Rabi seasons, maximum temperatures, precipitation etc.
6. Crop Prediction System using Machine Learning Algorithms, International Research
Journal of Engineering and Technology, Vol 7(2). The paper focuses on making use of
classification algorithms to improvise the crop yields. The paper is implemented using
Decision Tree Classifier, Naïve Bayes Classifier and KNN Classifier.

1.3 Organization of the Report

The remaining chapters of the project report are described as follows:

• Chapter 2 contains the proposed system, methodology, hardware and software details.
• Chapter 3 gives the cost involved in the implementation of the project.
• Chapter 4 discusses the results obtained after the project was implemented.
• Chapter 5 concludes the report.
• Chapter 6 consists of codes.
• Chapter 7 gives references.

8
 CHAPTER 2
AgriPredict: AI-Driven Crop Yield and Price Forecasting
This Chapter describes the proposed system, working methodology, software and hardware details.

2.1 Proposed System

The following block diagram (figure 1) shows the Work flow of this project.

Figure 2. System Block Diagram

9
2.2 Working Methodology

This System is a web-based application developed using HTML, CSS, and Flask . It allows farmers
to input essential details directly through a user-friendly interface. The inputs include information
such as land area both used and cultivated, crop type, fertilizers and nutrients used, and current
market prices. Once the farmer submits the data, the system processes the inputs and applies a
Random Forest Model, trained on historical data, to predict the crop yield. The system then
combines the yield prediction with real-time market prices and inflation rates to estimate the
farmer's expected income at the time of harvest. The output, which includes predicted crop yield,
estimated income, and market price, is displayed on the web interface. Additionally, the application
provides suggestions on the most profitable crops based on the given inputs, helping farmers make
informed decisions for the upcoming cultivating season. This web application aims to assist farmers
in optimizing their crop production and financial planning without the need for additional hardware
or complex setups.

2.3 Standards
Various standards used in this project are:
• HTTP(Hyper Text Transfer Protocol)
HTTPS ensures secure communication between the user's browser and the web server by
encrypting the data using SSL/TLS. This is crucial for protecting user data, particularly for any
personal or sensitive information that might be shared during the data entry process. It ensures that
the data transmitted between the client (farmer) and the server is encrypted, preventing
unauthorized access.
• W3C Web Standards (HTML5, CSS3, JavaScript ES6):
Your web application should follow the W3C web standards for HTML5, CSS3 These standards
ensure that the website is accessible, compatible across different browsers, and mobile-responsive.
HTML5 offers semantic elements, multimedia capabilities, and better accessibility, while CSS3
enables advanced styling techniques like animations, transitions, and responsive designs

10
2.4 System Details

This section describes the software and hardware details of the system:

2.4.1 Software Details

• WEB BASED APPLICATION (Website)

This is a web-based application built using HTML and CSS for the frontend, providing a user-friendly
interface for farmers to input data and view predictions. Flask is used for the backend, handling server-
side logic and integrating machine learning models for crop yield prediction. The application processes
user inputs, applies the trained machine learning model, and provides predictions along with market
price estimations. By leveraging Flask and web technologies, the system offers an efficient, accessible
solution for farmers to optimize their crop production and financial planning.

Developing Web application

The below are the snapshots of the project showing the functionality of the website

Fig 3:HOME PAGE

11
Fig 4:LOGIN PAGE

Fig 5.SIGN UP PAGE

12
 Fig 6.INPUT PAGE

Fig 7.Guide page

Fig 8.Help Page

13
 Fig 10:Output Page

2.4.2 Hardware Details

The Crop Yield Prediction System is entirely software-based and does not involve any hardware
components. The project is implemented as a web-based application using modern web
development technologies, including HTML, CSS, JavaScript for the frontend, and Flask for the
backend. It facilitates data collection directly from the user through a simple, intuitive interface.
Farmers input details such as land area, crop type, fertilizers, and other relevant factors, which are
then processed by a trained machine learning model to provide yield predictions and income
estimations. The system's focus is solely on software solutions, eliminating the need for additional
hardware, making it cost-effective and accessible to users with just an internet connection.

14
 CHAPTER 3
COST ANALYSIS

3.1

We do not use any hardware components with in our project. It is a web based application
The project is implemented as a web-based application using modern web development
technologies, including HTML, CSS, JavaScript for the frontend, and Flask for the backend. It
facilitates data collection directly from the user through a simple, intuitive interface. Farmers input
details such as land area, crop type, fertilizers, and other relevant factors, which are then processed
by a trained machine learning model(Random Forest) to provide yield predictions and income
estimations. The system's focus is solely on software solutions, eliminating the need for additional
hardware, making it cost-effective and accessible to users with just an internet connection. So it is
a ZERO COST APPLICATION.

TOTAL COST=Rs.0.

15
 CHAPTER 4

RESULTS AND DISCUSSIONS

4.1 Tentative Methodologies

• Problem Identification
• Data Collection
• Dataset Processing
• Data Visualization to gain insights
• Getting the data ready for training
• Training the machine learning models
• Finalizing the model that gives best crop yield
• Relating the crop yield with current and probable future market prices

4.2 Data Set information

The dataset is collected from official and trusted resources like FAO (Food and
Agricultural Organization of the United Nations) and Indian Government's data
resources like ODG (Open Government Data).
Multiple datasets are collected on the following categories:
• Crop yield
• Temperature data
• Fertilizers by nutrients containing Nitrogen, Potash, Phosphate
• Pesticides
• Quantity Produced
• Area Irrigated
• Area Harvested
• Land Used
• Synthetic Fertilizers

16
4.3 Processing Data Sets

• All the collected datasets have the data related to 80 different crops that were
grown in India spanning over 30 years (from 1990 to 2019).
• The columns in each dataset are checked thoroughly and unwanted features
are removed .
• Redundant features are removed by keeping the important one out of them
• All the datasets are finally merged into a bigger dataset based on the crop
items and the year of production.
• This new dataset is used for the further process.

4.4 Exploring the Data Sets

4.4.1 Distribution of Data:

Fig 10:Distribution of Data

17
4.4.2 Statistical Description

Table 1: Statistical Representation of data

4.5 Visualizing the data to find Insights

4.5.1 CORRELATION MATRIX

Table 2: Correlation Matrix

18
Fig 11. HEAT MAP

Fig 12. Scatter Plot

19
Fig 13.Bar Graphs

20
4.6 BUILDING PIPELINES FOR DATA PREPROCESSING

A sequence of data processing components is called a data pipeline. Pipelines are

very common in Machine Learning systems, since there is a lot of data to
manipulate and many data transformations to apply

4.6.1 Numerical Pipeline

I .Dealing with null values

• Sklearn's SimpleImputer
• Imputation transformer for completing missing values.
• Strategy = median: Replacing null values with median value of the column

II . Standard Scaling

• First it subtracts the mean value (so standardized values always have a zero mean), and then it divides by the
standard deviation so that the resulting distribution has unit variance.
• Standardize features by removing the mean and scaling to unit variance.
• The standard score of a sample x is calculated as: z = (x - u) / s

21
4.6.2 Categorical Pipeline

ONE HOT ENCODING

• Creates one binary attribute per category.

• This is called one-hot encoding, because only one attribute will be equal to 1
(hot), while the others will be 0 (cold).
• The new attributes are sometimes called dummy attributes.
• Scikit-Learn provides a OneHotEncoder class to convert categorical values
into one-hot vectors

Fig 14. One hot encoding

22
4.7 SELECTING AND TRAINING THE MACHINE LEARNING MODEL

• From the problem and the dataset, we can infer that this is a regression problem.
• Various regression algorithms are selected and trained to find the best model out of them.
• The models trained in this project are
i. Linear Regressor
ii. Decision Tree Regressor
iii. Random Forest Regressor
iv. Gradient Boosting Regressor
v. Linear Support Vector Regressor (SVR)

4.8 EVALUATION METRICS

• When it comes to regression algorithms, accuracy of the model doesn't work.

• It is because we cannot compare between real values which extend infinitely between the
limits.
• Hence for the regression algorithms, we try to find the error that a model makes.
• Root Mean Squared Error is one such evaluation metric, which calculates the error made
by the model on all instances.
• R2 Score helps in determining how better the model can perform by calculating the
proportion of variance in the dependent variable that is predictable from the independent
variables.

R2 SCORE:

ROOT MEAN SQUARED ERROR:

23
4.8.1

PERFORMANCE OF MODELS ON TRAINING SET

Table 2:Train R2 Score on Training set

ANALYSIS FROM THE TRAINING RESULTS

• Decision Tree Regressor has highest R2 Score = 1. But the RMSE value of Decision Tree
Regressor is worse when compared to the Random Forest Regressor.
• This means the Decision Tree Regressor has clearly overfitted the training data
• The Random Forest Regressor did well in terms of both RMSE value and R2 Score.
• The Linear SVR did not perform well and clearly it shows a negative R2 Score, this means
the Linear SVR is not following the trend in the dataset and it is fitting worse than a
horizontal line.
• The Linear Regressor fitted the data well, but has poor performance when compared to
Random Forest Regressor and Decision Tree Regressor
• The Gradient Boosting Regressor's weak decision trees couldn't perform well as compared
to individual Decision Tree and Random Forest.

24
4.8.2

EVALUATION OF MODELS ON THE TEST SET

Table 3:Test R2 Score on Test set

ANALYSIS FROM THE TESTING RESULTS

• The decreased R2 Score of Decision Tree Regressor on test set as compared

to training set clearly shows that the model has overfitted the data.
• The Random Forest Regressor performed well on both training and testing
sets in terms of RMSE and R2 scores
• The Linear Regressor has relatively less scores as compared to the Random
Forests.
• The Gradient Boosting Regressor and the Linear SVR hasn't shown good
performance on the testing set too.
• Random Forest Regressor's ensemble nature makes it the most powerful
machine learning algorithm. The combined predictions of all the individual
model's in the ensembled Random Forest clearly gave better results.

25
4.9 COMBINING YIELD WITH PRICES

• The crop yield predicted by the machine learning model will be combined
with the market prices and inflation at that day to predict the amount that the
farmer is going to get at the time of harvest.
• The farmer inputs the required fields in a small farmer friendly UI, which
then processes the whole calculation and shows the output.

26
 CHAPTER 5

CONCLUSION AND FUTURE WORK

• This project depends on the input given by the farmer based on the current
price of the crop. This can be improved by collecting data on crop prices
across the country.
• This project can be specialized regionally (for example, statewide crop yield
prediction) depending on the factors prevailing in that region.
• More climatic factors like rainfall, precipitation, wind speed, humidity, etc
can also be taken into consideration in order to improve the model’s
efficiency in terms of climatic conditions.
• Data related to the soil present in the crop can also be considered for more
insights by examining the soil through other artificial intelligence techniques.
• Considering the effects of the natural disasters might help the farmers in
exceptional environmental conditions.
• More crops can be added to the list of the crops in order to increase the reach
of the project to more farmers.
• We can create a Web Application to get inputs from farmers.
• The application’s UI can be made more farmer-friendly by supporting
regional languages.

27
 CHAPTER 6

APPENDIX

Back-End Flask Code

from flask import Flask, render_template, request, redirect, url_for, session, flash

import pickle

import pandas as pd

app = Flask(_name_)

app.secret_key = 'your_secret_key' # Replace with a secure secret key

# In-memory user data for demonstration (in production, use a database)

authorized_users = {"user1": "password1", "user2": "password2"}

# Load the model pipeline and regression model

with open('models/full_pipeline_model.sav', 'rb') as f:

full_pipeline = pickle.load(f)

with open('models/rf_regmodel.sav', 'rb') as f:

model = pickle.load(f)

# Homepage Route

@app.route('/')

def homepage():

return render_template('homepage.html')

# Login Page Route

@app.route('/login', methods=['GET', 'POST'])

def login():

if request.method == 'POST':

username = request.form.get('username')

password = request.form.get('password')

# Simple authentication check

if username in authorized_users and authorized_users[username] == password:

session['logged_in'] = True

session['username'] = username

return redirect(url_for('main_index'))

else:

flash('Invalid credentials. Please try again.')

return redirect(url_for('login'))

return render_template('login.html')

28
# Signup Route (if needed for adding new users)

@app.route('/signup', methods=['GET', 'POST'])

def signup():

if request.method == 'POST':

username = request.form.get('username')

password = request.form.get('password')

# Register new user if username doesn't already exist

if username in authorized_users:

flash('Username already exists. Please choose a different one.')

return redirect(url_for('signup')

authorized_users[username] = password

flash('Signup successful! Please log in.')

return redirect(url_for('login'))

return render_template('signup.html')

# Guide Page Route

@app.route('/guide')

def guide_page():

return render_template('Guide.html')

# Help Page Route

@app.route('/help')

def help_page():

return render_template('help.html')

# Main Functionality Page (requires login)

@app.route('/index')

def main_index():

if not session.get('logged_in'):

flash('Please log in to access this page.', 'error')

return redirect(url_for('login'))

return render_template('index.html')

# Logout Route

@app.route('/logout')

def logout():

session.pop('logged_in', None)

flash('You have been logged out.')

return redirect(url_for('homepage'))

# Prediction Submission Route - Requires Login

@app.route('/submit', methods=['POST'])

def submit():
29
 if not session.get('logged_in'):

return redirect(url_for('login'))

# Process form data for prediction

user_input = {

"CropItem": request.form.get("CropItem"),

"Year": int(request.form.get("Year")),

"Nutrients(tonnes)": float(request.form.get("Nutrients(tonnes)")),

"SyntheticFert(tonnes)": float(request.form.get("SyntheticFert(tonnes)")),

"Pesticides(tonnes)": float(request.form.get("Pesticides(tonnes)")),

"Temp_ann_degC": float(request.form.get("Temp_ann_degC")),

"LandUsed": float(request.form.get("LandUsed")),

"LandIrrigated": float(request.form.get("LandIrrigated")),

"MarketPrice": float(request.form.get("MarketPrice")),

try:

yield_value, future_income = get_predictions(user_input)

input_values = list(user_input.values())

return render_template('submit.html', input_values=input_values, yield_value=yield_value, future_income=future_income)

except Exception as e:

flash(f"Prediction error: {e}")

return redirect(url_for('main_index'))

# Helper Functions for Prediction

def get_processed_data(user_input):

user_input = convert_units(user_input)

user_df = pd.DataFrame([user_input])

return full_pipeline.transform(user_df)

def get_predictions(user_input):

market_price = user_input["MarketPrice"]

crop_area = user_input["LandUsed"]

del user_input["MarketPrice"]

processed_data = get_processed_data(user_input)

yield_value = model.predict(processed_data)[0] / 24.17

predicted_price = (100 + get_inflation_rate()) * market_price / 100

future_income = round(yield_value * predicted_price * crop_area, 2)

return round(yield_value, 3), future_income

def convert_units(user_input):

user_input["Nutrients(tonnes)"] /= 1000
30
 user_input["SyntheticFert(tonnes)"] /= 1000

user_input["Pesticides(tonnes)"] /= 1000

user_input["LandUsed"] /= 2471.052

user_input["LandIrrigated"] /= 2471.052

return user_input

def get_inflation_rate():

try:

inflation = pd.read_csv("datasets/InflationRates.csv")

inflation["Value"].fillna(inflation["Value"].median(), inplace=True)

inflation = inflation.groupby("Year")["Value"].mean().reset_index()

return round(inflation["Value"].median() / 2, 5)

except FileNotFoundError:

flash("Inflation data file missing.")

return 0

# Run the App

if _name_ == '_main_':

app.run(debug=True)

FRONT END HTML AND CSS CODES:

Home page

31
32
33
LOGIN PAGE

34
HELP PAGE

35
36
Training the model over the data set

37
38
39
40
41
42
43
 REFERENCES
• S. Veenadhari, B. Misra and C. Singh, "Machine learning approach for forecasting crop yield
based on climatic parameters," 2014 International Conference on Computer Communication
and Informatics, Coimbatore, 2014, pp. 1-5, doi: 10.1109/ICCCI.2014.6921718.
• Balakrishnan, N., Muthukumarasamy, G.: Crop production-ensemble machine learning model
for prediction. Int. J. Comput. Sci. Softw. Eng. 5(7), 148–153 (2016)
• Jeong JH, Resop JP, Mueller ND, Fleisher DH, Yun K, Butler EE, Timlin DJ, Shim KM,
Gerber JS, Reddy VR, Kim SH. Random Forests for Global and Regional Crop Yield
Predictions. PLoS One. 2016 Jun 3;11(6):e0156571. doi: 10.1371/journal.pone.0156571.
PMID: 27257967; PMCID: PMC4892571.
• Mishra, Subhadra & Mishra, Debahuti & Santra, Gour. (2016). Applications of Machine
Learning Techniques in Agricultural Crop Production: A Review Paper. Indian Journal of
Science and Technology. 9. 10.17485/ijst/2016/v9i38/95032.
• Priya, P., Muthaiah, U., & Balamurugan, M. (2018). Predicting yield of the crop using
machine learning algorithm. International Journal of Engineering Sciences & Research
Technology, 7(1), 1-7.
• Pavan P., Virendra P., Shrikhant K., Crop Prediction System using Machine Learning
Algorithms, International Research Journal of Engineering and Technology (IRJET), 7(2),
2020 Feb.
• Aurelien Geron, Hands-on Machine Learning with Scikit-Learn, Keras and TensorFlow, 2nd
Edition, O’Reilly.
• David Freedman, Robert Pisani, Roger Purves, Statistics, 4th Edition, Viva Books Pvt ltd.
• Wes McKinney, Python for Data Analysis, 2nd Edition, O’Reilly.
• Jake VanderPlas, Python Data Science Handbook, O’Reilly.
• Peter Harrington, Machine Learning in action, Manning Publications
• Samprit Chatterjee, Ali S. Hadi, Regression Analysis by Example, 5th Edition, Wiley.

44
 BIODATA
TEAM MEMBER-1

Name: Neelam.Suvarna Jyothi

Mobile :8367382676
Email: jyothi.21bce9552@vitapstudent.ac.in
Address: 2 -42,katlampudi,Eluru Mandal,Eluru District

TEAM MEMBER -2

Name: Parepalli Dhyanesh

Mobile : 9603972166
Email: dhyanesh.21bce9397@vitapstudent.ac.in
Address: adamilli,Kamavarapukota mandal,Eluru District

TEAM MEMBER -2

Name: Kondapalli Vasavi

Mobile : 6304267080
Email: vasavi.21bce9474@vitapstudent.ac.in
Address: Kothapet,Guntur

TEAM MEMBER -4

Name: Shaik Rokiya

Mobile : 8125830064
Email: rokiya.21bce9546@vitapstudent.ac.in
Address: Mangaldasnagar,Guntur

45