Professional Documents
Culture Documents
Flipkart Reviews Sentiment Analysis
Flipkart Reviews Sentiment Analysis
on
Submitted to
KIIT Deemed to be University
BACHELOR’S DEGREE IN
COMPUTER SCIENCE AND TECHNOLOGY
BY
Submitted to
KIIT Deemed to be University
In Partial Fulfilment of the Requirement for the Award of
BACHELOR’S DEGREE IN
COMPUTER SCIENCE AND
ENGINEERING
BY
CERTIFICATE
This is certify that the project entitled
“Flipkart Reviews Sentiment Analysis“
submitted by
is a record of bonafide work carried out by them, in the partial fulfilment of the
requirement for the award of Degree of Bachelor of Engineering (Computer Sci-
ence & Engineering OR Information Technology) at KIIT Deemed to be university,
Bhubaneswar. This work is done during year 2022-2023, under our guidance.
Date: / /
We are profoundly grateful to Dr. Murari Mandal of Affiliation for his expert
guidance and continuous encouragement throughout to see that this project rights its
target since its commencement to its completion. .....................
3 Problem Statement 4
3.1 Project-Planning 4
3.2 Project-Analysis 4
3.3 System-Design 5
3.3.1 Design-Constraints 5
3.3.2 Block-Diagram 5
4 Implementation 6
4.1 Methodology or Proposal 6
4.2 Testing 6
4.3 Result Analysis and Screenshots 7
4.4 Quality Assurance 7
5 Standard Adopted 8
5.1 Standards of Design 8
5.2 Standards of Coding 8
5.3 Standards of Testing 8
References 10
Individual Contribution 11
Plagiarism Report 12
List of Figures
Chapter 1
Introduction
Although there are many existing sentiment analysis tools available, they often
face challenges in accurately identifying the sentiment of a text, especially with
the nuances of human language. This project aims to address these gaps by
developing a machine learning model that can improve the accuracy of sentiment
analysis.
Chapter 2
2. Basic Concepts/Literature-Review
2.1 Natural Language Processing (NLP)
Within NLP, sentiment analysis is a subfield that deals with detecting and
extracting subjective information from text, including attitudes, emotions, and
opinions. Sentiment analysis has many applications, including brand monitoring,
customer feedback analysis, and political analysis.
Text classification is the task of categorizing text documents into one or more
predefined categories. It is a common NLP task that has many applications, such
as spam filtering, topic modeling, and sentiment analysis. Text classification
algorithms use various features, such as word frequencies and n-grams, to
classify text.
Data mining is the practice of analyzing large datasets to uncover patterns and
insights, employing statistical and machine learning methods to identify
relationships and patterns. Data mining has many applications, including
marketing, fraud detection, and recommendation systems.
Social media analysis involves analyzing data from social media platforms, such
as Twitter and Facebook, to gain insights into user behavior and opinions. Social
media analysis has many applications, including brand monitoring, market
research, and political analysis.
Chapter 3
Problem Statement
The project report aims to analyze the effectiveness of a new marketing
strategy implemented by a company. The study will involve collecting data on
the sales performance of the company before and after the implementation of
the strategy. The data will be analyzed using statistical tools to determine the
impact of the new marketing approach on the company's sales. The report will
provide insights into the effectiveness of the strategy and recommendations for
future marketing efforts.
The problem statement is to develop a sentiment analysis model for social media
data. After analyzing the problem statement, the following requirements have
been identified:
The model should be able to handle large datasets of social media text data
The model should be able to classify the sentiment of the text data as
positive, negative, or neutral
The model should be able to handle various types of text data, including
short and informal text data
3.3System Design
3.3.1Design Constraints
Chapter 4
Implementation
In this section, the implementation of the proposed sentiment analysis model is
presented.
4.2 Testing
To evaluate the performance of the proposed model, it is tested on a dataset of
customer reviews from various online platforms. The testing is done using a 10-
fold cross-validation technique, and the evaluation metrics used are accuracy,
precision, recall, and F1-score. The testing results are presented in the form of a
confusion matrix and performance metrics.
Test Test Case Title Test Condition System Behavior Expected Result
ID
System Analyses
the text and classify
User provides a it as positive
T01 Positive Review positive review sentiment Positive
System Analyses
the text and classify
User provides a it as Negative
T02 Negative Review Negative review sentiment Negative
System Analyses
the text and classify
User provides a it as Neutral
T03 Neutral Review Neutral review sentiment Neutral
4.3 Result Analysis OR Screenshots
The results obtained from the testing and evaluation of the sentiment analysis
model are presented in this section. The performance metrics of the model show
that it achieves a high accuracy of 86% and F1-score of 0.86, indicating that the
model is effective in classifying the sentiment of customer reviews. Screenshots
of the model's output are also provided.
Chapter 5
Standards Adopted
Chapter 6
6.1 Conclusion
References
https://www.analyticsvidhya.com/
https://www.kaggle.com/
Wikipedia
GeeksForGeeks
Abstract: This project aims to conduct sentiment analysis on customer reviews of products
sold on Flipkart, a leading e-commerce platform in India, using Python and data science
libraries. By analyzing sentiments, companies can understand customers' opinions and
enhance their products and services. The project involves data cleaning, applying NLTK
Sentiment Vader to analyze reviews, and data visualization techniques like word clouds and
bar plots. The project offers a practical experience of applying data science techniques to real-
world problems in e-commerce.
Individual contribution and findings: My contribution to the Flipkart reviews
sentiment analysis project focused on analyzing the results of the NLTK Sentiment Vader
model. I used data visualization techniques such as bar plots and word clouds to identify
patterns in the data and interpret the sentiments expressed in the reviews. My findings showed
that the majority of reviews had positive sentiment, with common words like "good," "great,"
and "excellent." Negative reviews used words such as "poor," "worst," and "bad." These
insights could help e-commerce companies like Flipkart improve their products and services.
My analysis contributed to a better understanding of the data and its implications for the
project.