National Education Society ®

JNN College of Engineering

Sawalanga Road, Navule, Shivamogga -277204
(Accredited by NBA, NAAC 'B' & Certified by UGC 2f & 12B)
Department of Computer Science Engineering
INTERNSHIP SEMINAR
On
“FLIPKART REVIEW SENTIMENTAL ANALYSIS”
Carried out at
10 SECONDS co.in, BANGALORE
during
22.08.2022 TO 14.09.2022

Presented by: External Guide: Internal Guide:

Abhishek k Dr . H C Ramaprasad Prof. Madhusudhan G
About the organization

• 10 SECONDS is an organization that unifies training, assessments and

recruitment under one roof. Conceptualized by Anijith Shetty and
Ramaprasad, they are an ensemble set of young executive trainers
who specialise in all things digital.
• The name of the business is inspired by research that suggests it takes
10 SECONDS to build a strong first impression. They try to apply this
principle in everything they do, be it in our interactions with clients,
prospects or even our partners.
• They focus on technology and innovation makes us future-ready and
helps us deliver cutting edge solutions that are student-centric, easy
to use and most importantly cost-effective for our clients.
Dep OF t E&TE
07/05/2024 2

Area of Internship
1. C Programming
• C is a procedural programming language
initially developed by Dennis Ritchie in the
year 1972 at Bell Laboratories of AT&T
Labs.
• It was mainly developed as a system
programming language to write the UNIX
operating system.
Dep of E&TE
07/05/2024
2. C++ Programming Language
• C++ is a general-purpose compiled
programming language that was
developed as an enhancement of the C
language to include object-oriented
paradigm.
• The basic syntax and code structure of
both C and C++ are the same.
3
 Web Development
• Web development is the work involved in developing a website for the Internet (World
Wide Web) or an intranet (a private network)
• Web development refers to the creating, building, and maintaining of websites . It is
the creation of an application that works over the internet i.e. websites.
• Web Development can be classified into two ways:
 Frontend Development.
 Backend Development.

Dep of E&TE
07/05/2024 4
 SOFTWARE IMPLIMENTATION
1. GOOGLE COLAB
• Google colab, is a product from Google Research. Colab allows anybody to write and execute arbitrary
python code through the browser.
• Google Colab is simply an online representation of Jupyter Notebook. While Jupyter Notebook needs
installation on a computer and can only use local machine resources, Colab is a full-fledged cloud app
for Python coding.
• Colab is a data analysis and machine learning tool that allows you to combine executable Python code
and rich text along with charts, images, HTML, LaTeX and more into a single document stored in
Google Drive.

Dep of E&TE
07/05/2024 5
2. PYTHON
• Python is an interpreted, interactive, object-oriented programming language. it is
designed by Guido van Rossum.
• It supports multiple programming paradigms beyond object-oriented
programming, such as procedural and functional programming.
• Python works on different platforms (Windows, Mac, Linux, Raspberry Pi, etc)
and has a simple syntax similar to the English language.

Dep of E&TE

07/05/2024 Dep of E&TE 6

INTRODUCTION
• Flipkart is one of the leading e-commerce companies/platforms based in India, which
currently competes with Snapdeal, Myntra, Nykaa, and IndiaMART. Since the world is
digitizing rapidly, e-commerce platforms are booming and making the best way possible
to make and provide their services to the people.
• The sentiment analysis of the product reviews is a way that can open ways to
understand the performance of the products, which is beneficial for both future
customers and the e-commerce platform.
• The e-commerce company can improve its product and customer services with the
sentiment analysis technique. Sentiment analysis can also help the company better
understand the product and plan further regarding products.

Dep of E&TE
07/05/2024 7
AIM OF THE PROJECT
1. The aim of this analysis is to identify customer sentiment about the
product or service, and to gain insights into customer experience with
Flipkart.
2 . This analysis can be used for decision-making, improve customer
service, and to measure customer satisfaction.

Dep of E&TE
07/05/2024 8
THEORETICAL BACKGROUND
• The origin of sentiment analysis can be traced to the 1950s, when
sentiment analysis was primarily used on written paper documents.
• Product reviews can be analyzed to understand the sentiment of the
people towards a particular topic. therefore a summary of positive
and negative reviews needs to be generated.
• Here, the main focus is on the review of algorithms and techniques
used for extract feature wise summary of the product and analyzed
them to form an authentic review.

Dep of E&TE
07/05/2024 9
 FLOW CHART

Dep of E&TE
07/05/2024 10
DESIGN AND IMPLEMENTATION

• Sentiment analysis refers to analyzing an opinion or feelings about

something using data like text or images. Sentiment analysis helps
companies in their decision-making process
• Sentiment analysis, an important area in NLP, is the process of
automatically detecting affective states of text. It can be just a basic
task of classifying the polarity of a text as being positive/negative or it
can go beyond polarity, looking at sentiment states etc.

Dep of E&TE
07/05/2024 11
 WORKING PRINCIPLE
1. Importing the libraries like pandas, ntlk, matplotlib, wordcloud , RegEx, nltk
sentiment Vand Stopwords. The dataset consists of reviews of the products which
customers purchased from Flipkart. The dataset for Flipkart contains 3 columns:
• Product_name: the name of the product with little details about it.
• Review: review what the customer gives about that product.
• Rating: ranges from 1 to 5 stars given by the customer.
2. The sentimental analysis task begins if the dataset does not have any null values.
So, we have to check whether the dataset contains any null values or not.
3. Then The data cleaning of the review column is done with the use of the regular
expression, stemming, and stopwords. Here, stemmar is simply stemming, which
converts the words into their root form and the wordcloud is used to view the most
common or frequent words used by customer to give a product review.

Dep of E&TE
07/05/2024 12
4. After cleaning the data, we need to see how most people rate the products they purchase
from Flipkart. The rating range begins from 1 to 5.
5. The total counts of ratings can be plotted in the form of a pie chart to give us a visual
representation and proportion of each rating category.
6. To view the most common or frequent words used by customers to give a product review.
The Wordcloud is used to visualize the most frequently used words in a bigger font. This
is used for the reviews column. WordCloud visualizes the data in a textual representation.
7. In WordCloud, the most frequently used words or important words will be depicted in
large font sizes, and the less frequently used words will have a smaller font size.
8. Sentimental analysis requires review’s sentiment score. By this it can predicted what that
review actually means, and it can help a future customer who is willing to purchase that
product as well as the company to understand that product.
9. For this, We have to add 3 columns to the dataset, which are positive, negative and
neutral, which will calculate the sentiment score of each review. Also, the sentiment score
will be decided based on the polarity score that determines under what category the
review comes.
10. Bifurcate and print the sentiment score of positive, negative and neutral reviews to know
the exact number of reviews in each category.
Dep of E&TE
07/05/2024 13
RESULTS

So, by the above output, there are 923.55 Positive reviews, 96.77 Negative reviews,
and 1283.68 Neutral reviews. With the final output of the entire implementation, the
key takeaways can be stated as follows:
Most of the reviews are in favor of Neutral.
The Negative review score has a small proportion, and Positive reviews are
midway.
So with the help of the above sentiment analysis implementation, we can conclude
that most people are satisfied with the products
07/05/2024
Dep of E&TE and services they received from
14
CONCLUSION
• The aim of this project is to analyze the sentiment reviews of the Flipkart
products. So, by the final output of the project, there are
 923.55 positive reviews,
 96.77 negative reviews
 1283.68 neutral reviews.
• Most people give Neutral reviews, and a small proportion of people give
Negative reviews. This states that sentiment analysis is a good process to
gather insights into each product, which ultimately benefit present and future
customers and e-commerce companies.
• Sentiment analysis is important because, based on bad reviews, the e-commerce
company makes those products better or replaces those products with better and
newer ones, which ultimately improves the overall customer service too.
Dep of E&TE
07/05/2024 15
OUTCOME OF THE INTERNSHIP
The internship in C & C++ programming help us to develop technical
skills in programming and software development, as well as gain
experience working on real-world software projects. From the web
development we learnt technical skills, such as HTML, CSS,
JavaScript, and other web technologies. we also gained experience in
web design, user experience (UX), and web development frameworks
and tools.

07/05/2024 Dep of E&TE 16

 REFERENCES
1. “Reviewing flipkart product comments using methods based on sentiment analysis
(2023)” by adiya rohanroy ketan rathor, bhavesh jannawat , aditya, prof. S.C. Sethi
pune-41, maharashtra, india.
2. "Sentiment Analysis of Movie Reviews Using R (2019).“by Anandarajan, Murugan,
Chelsey Hill, and Thomas Nolan Practical Text Analytics. Springer, Cham.
3. "Sentiment Analysis on WhatsApp Group Chat Using R. (2019)" by Joshi, Sunil. Data,
Engineering and Applications. Springer, Singapore, 47.
4. "Survey on clustering techniques in Twitter data(2020).“by Devika, R., and S.
Revathy. Second International Conference on Computing Methodologies and
Communication (ICCMC). IEEE, 1073-1077.
5. "Analyzing Airbnb Customer Experience Feedback Using Text Mining(2019)“.by
Joseph, George, and Vinu Varghese. Big Data and Innovation in Tourism, Travel, and
Hospitality. Springer, Singapore, 147-162.
Dep of E&TE
07/05/2024 17