“SENTIMENT ANALYSIS WEB

APPLICATION”

A Synopsis submitted to
JSS Science and Technology University
in partial fulfilment for the award of the degree of

Master of Science
in

Data Science
by
Tejaswini K S (01JST23PCD005)

Under the supervision of

DR. Prasanna B T
Associate Professor
Department of Computer Science
JSS STU, Mysuru-06

Department of Information Science and Engineering

JSS Science and Technology University
2023-24
 Abstract

This project presents the development of an interactive sentiment analysis application

using Streamlit in combination with Python libraries TextBlob, Pandas, and Cleartext.
TextBlob is utilized for its straightforward sentiment classification capabilities, analyzing text
to determine emotional polarity and subjectivity. Pandas aids in managing and preprocessing
text data, ensuring efficient data handling and integration. Cleartext enhances text quality by
performing noise reduction and normalization. Streamlit provides a dynamic web interface for
real-time sentiment analysis, allowing users to input text and instantly view sentiment results
through an intuitive and interactive dashboard. This integration of tools creates an accessible
and user-friendly platform for evaluating and visualizing textual sentiments, making advanced
sentiment analysis more approachable and engaging.
 Chapter 1 Introduction

1.1 Introduction
Sentiment analysis is a crucial technique in natural language processing (NLP) that
involves interpreting and classifying the emotional tone of text data. It has applications in
various fields, including customer feedback, social media monitoring, and market research.
The increasing volume of textual data from online sources necessitates efficient and user-
friendly tools to analyze sentiments effectively.
This project aims to develop an interactive sentiment analysis application using Python
libraries and the Streamlit framework. TextBlob is employed for its ease of use in extracting
sentiment polarity and subjectivity from text. Pandas is used for data manipulation and
preprocessing, ensuring that the input data is well-structured for analysis. Cleartext assists in
refining text data by performing essential preprocessing tasks such as noise reduction and
normalization.
Streamlit serves as the platform for building an intuitive web interface, allowing users
to interact with the sentiment analysis tool in real-time. This combination of technologies
provides a seamless and engaging user experience, facilitating quick and accurate sentiment
assessment. By integrating these tools, the project aims to make sentiment analysis more
accessible and interactive, enabling users to gain valuable insights from textual data
effortlessly.

1.2 About the Project

This project involves the creation of an interactive sentiment analysis application
designed to evaluate the emotional tone of textual data. The core objective is to develop a
user-friendly tool that leverages Python libraries and the Streamlit framework to provide
real-time sentiment insights.
The application integrates several key components:
1. TextBlob: Utilized for its straightforward sentiment analysis capabilities, TextBlob
classifies text into various sentiment categories—positive, negative, or neutral—
based on its polarity and subjectivity scores.
2. Pandas: Employed for efficient data handling and preprocessing, Pandas manages
the text data, ensuring it is in a suitable format for analysis and facilitates seamless
data integration.
 3. Cleartext: Used to enhance the quality of the text data by performing essential
preprocessing tasks, such as removing noise and normalizing text, which improves
the accuracy of the sentiment analysis.
4. Streamlit: Provides the interactive web interface where users can input text and
receive instant sentiment feedback. Streamlit’s real-time capabilities and intuitive
design make the sentiment analysis process engaging and accessible.
The project’s goal is to offer an interactive, easy-to-use platform that enables users to
quickly assess and visualize the sentiment of textual inputs. By combining these
technologies, the application aims to streamline sentiment analysis, making it more practical
for various real-world applications such as customer feedback analysis, social media
monitoring, and more.

1.2.1 Motivation

The exponential growth of digital communication through social media, customer

reviews, and online forums has generated vast amounts of textual data, necessitating efficient
tools for extracting meaningful insights. Traditional methods of sentiment analysis often
involve complex and inaccessible systems, which can be a barrier to users seeking real-time
and actionable feedback.
This project is motivated by the need to democratize access to sentiment analysis by
creating an intuitive, interactive tool that simplifies the process of understanding textual
sentiment. By leveraging Python libraries and the Streamlit framework, the project aims to
bridge the gap between advanced sentiment analysis techniques and everyday users. The goal
is to provide a user-friendly application that not only performs robust sentiment analysis but
also presents results in an engaging and accessible format.
The motivation extends to enhancing user experience through real-time feedback and
interactive visualizations, which can be valuable for businesses, researchers, and individuals
seeking to make informed decisions based on text data. By integrating state-of-the-art NLP
tools with an easy-to-use interface, this project seeks to make sentiment analysis more
approachable, practical, and impactful in a variety of real-world contexts.

1.2.2 Problem Definition

The rapid proliferation of textual data from sources such as social media, customer
reviews, and online forums presents a significant challenge in effectively analyzing and
 interpreting sentiment. Traditional sentiment analysis tools often suffer from complexity,
lack of real-time feedback, and limited user engagement, making them less accessible to non-
experts.
The primary problem addressed by this project is the need for a user-friendly,
interactive tool that simplifies sentiment analysis and makes it readily accessible. Users
require a solution that not only performs accurate sentiment classification but also presents
results in an intuitive and engaging manner. Existing methods may involve cumbersome
processes or technical barriers, limiting their practical use for individuals and organizations
seeking immediate insights from textual data.
This project seeks to overcome these challenges by developing an application that
integrates Python libraries and Streamlit to deliver real-time sentiment analysis. The goal is
to provide an easy-to-use interface that allows users to input text and obtain clear, actionable
sentiment feedback, thereby bridging the gap between advanced sentiment analysis
techniques and practical, everyday use.

1.2.3 Challenges

1. Accuracy of Sentiment Analysis: Ensuring that the sentiment analysis tool

accurately classifies text into appropriate sentiment categories (positive, negative,
neutral) is challenging. This requires effective use of libraries like TextBlob and
proper preprocessing to handle nuances in language and context.
2. Data Preprocessing: Preparing text data for analysis involves several steps,
including noise reduction, text normalization, and handling diverse input formats.
The integration of Cleartext for preprocessing must be effective in improving data
quality without introducing errors.
3. Real-time Performance: Developing a system that processes text and provides
sentiment feedback in real-time can be demanding. Ensuring that the Streamlit
application performs efficiently and responds promptly to user input requires
optimizing both the backend processing and the user interface.
4. User Experience: Designing an intuitive and engaging interface with Streamlit that
effectively communicates sentiment analysis results poses a challenge. The
application must present data in a user-friendly manner, providing clear visualizations
and feedback to enhance user interaction.
 5. Integration of Tools: Seamlessly integrating multiple libraries (TextBlob, Pandas,
Cleartext) with Streamlit requires careful coordination to ensure that data flows
smoothly between components and that the overall system functions cohesively.
6. Handling Diverse Text Inputs: The tool must be robust enough to handle a wide
variety of text inputs, including slang, abbreviations, and domain-specific language,
which can affect the accuracy and reliability of sentiment analysis.
Addressing these challenges is crucial for developing a reliable, efficient, and user-friendly
sentiment analysis application.

1.2.4 Objectives

1. Develop an Interactive Interface: Create a user-friendly web application using

Streamlit that allows users to input text and receive real-time sentiment analysis
results.
2. Implement Accurate Sentiment Analysis: Utilize TextBlob to accurately classify
text into sentiment categories—positive, negative, or neutral—by analyzing the text's
polarity and subjectivity.
3. Enhance Data Quality: Employ Cleartext for effective text preprocessing, including
noise reduction and normalization, to improve the accuracy of sentiment analysis.
4. Ensure Efficient Data Handling: Use Pandas for managing and preprocessing text
data to facilitate smooth integration and analysis within the application.
5. Optimize Real-Time Performance: Design the application to process and analyze
text quickly, providing instant feedback and maintaining high performance.
6. Provide Clear Visualizations: Develop intuitive visualizations within the Streamlit
interface to present sentiment analysis results in a comprehensible and engaging
manner.

1.2.5 Methodology

1. Text Data Collection and Preparation:

o Gather and preprocess textual data from various sources to ensure it is suitable
for sentiment analysis. This involves cleaning the text, removing noise, and
normalizing content using the Cleartext library.
2. Sentiment Analysis Implementation:
o Integrate TextBlob to perform sentiment analysis on the preprocessed text.
TextBlob will classify the text into sentiment categories (positive, negative,
neutral) based on its polarity and subjectivity scores.
3. Data Management:
o Use Pandas for handling and organizing text data. This includes data loading,
manipulation, and integration to ensure that the input text is processed
efficiently for analysis.
4. Application Development:
o Develop the web application interface using Streamlit. Design and implement
features that allow users to input text and view sentiment analysis results in
real-time. Incorporate interactive elements and visualizations to enhance user
experience.
5. Integration and Testing:
o Integrate TextBlob, Pandas, Cleartext, and Streamlit within a cohesive
application framework. Conduct thorough testing to ensure the system
performs reliably, processes text accurately, and provides timely feedback.
6. Optimization and Refinement:
o Optimize the application for performance, ensuring that it processes and
analyzes text efficiently. Refine the user interface to improve usability and
engagement based on user feedback.
7. Deployment and Evaluation:
o Deploy the application for use and evaluate its effectiveness in providing
accurate sentiment analysis. Collect user feedback and performance metrics
to make further improvements as needed.
 Chapter 2 Literature Survey

[1] Saad and Yang, "Twitter Sentiment Analysis Based on Ordinal Regression," have
aimed for giving a complete tweet sentiment analysis on the basis of ordinal regression with
machine learning algorithms. The suggested model included pre-processing tweets as first
step and with the feature extraction model, an effective feature was generated. The methods
such as SVR, RF, Multinomial logistic regression (SoftMax), and DTs were employed for
classifying the sentiment analysis. Moreover, twitter dataset was used for experimenting the
suggested model. The test results have shown that the suggested model has attained the best
accuracy, and also DTs were performed well when compared over other methods.

[2] Y. Fang, H. Tan and J. Zhang, "Multi-Strategy Sentiment Analysis of Consumer

Reviews Based on Semantic Fuzziness," have suggested multi-strategy sentiment analysis
models using the semantic fuzziness for resolving the issues. The outcomes have
demonstrated that the proposed model has attained high efficiency.

[3] M. Afzaal, M. Usman and A. Fong, "Tourism Mobile App With Aspect-Based
Sentiment Classification Framework for Tourist Reviews," have recommended a novel
approach of aspect-based sentiment classification, which recognized the features in a precise
manner and attained the best classification accuracy. Moreover, the scheme was developed
as a mobile application, which assisted the tourists in identifying the best hotel in the town,
and the proposed model was analyzed using the real-world data sets. The results have shown
that the presented model was effective in both recognition as well as classification.

[4] A. Feizollah, S. Ainin, N. B. Anuar, N. A. B. Abdullah and M. Hazim, "Halal

Products on Twitter: Data Extraction and Sentiment Analysis Using Stack of Deep Learning
Algorithms," have concentrated on tweets related to two halal products such as halal
cosmetics and halal tourism. By utilizing Twitter search function, Twitter information was
extracted, and a new model was employed for data filtering. Later, with the help of deep
learning models, a test was performed for computing and evaluating the tweets. Moreover,
for enhancing the accuracy and building prediction methods, RNN, CNN, and LSTM were
employed. From the outcomes, it was seemed that the combination of LSTM and CNN
attained the best accuracy.
 [5] NeelamMukhtar, Mohammad AbidKhan, and NadiaChiragh, "Lexicon-based
approach outperforms Supervised Machine Learning approach for Urdu Sentiment Analysis
in multiple domains" have performed the sentiment analysis to the Urdu blogs attained from
several domain with Supervised Machine learning and Lexicon-based models. In Lexicon-
based models, a well-performing Urdu sentiment analyzer and an Urdu Sentiment Lexicons
were employed, whereas, in Supervised Machine learning algorithm, DT, KNN, and SVM
were employed. The data were combined from the two soruces for performing the best
sentiment analysis. Based on the tests conducted, the outcomes were shown that the Lexicon-
based model was superior to the supervised machine learning algorithm
 References

[1] S. E. Saad and J. Yang, "Twitter Sentiment Analysis Based on Ordinal Regression," IEEE
Access, vol. 7, pp. 163677-163685, 2019.

[2] Y. Fang, H. Tan and J. Zhang, "Multi-Strategy Sentiment Analysis of Consumer Reviews
Based on Semantic Fuzziness," IEEE Access, vol. 6, pp. 20625-20631, 2018.

[3] M. Afzaal, M. Usman and A. Fong, "Tourism Mobile App With Aspect-Based Sentiment
Classification Framework for Tourist Reviews," IEEE Transactions on Consumer
Electronics, vol. 65, no. 2, pp. 233-242, May 2019.

[4] A. Feizollah, S. Ainin, N. B. Anuar, N. A. B. Abdullah and M. Hazim, "Halal Products on

Twitter: Data Extraction and Sentiment Analysis Using Stack of Deep Learning Algorithms,"
IEEE Access, vol. 7, pp. 83354-83362, 2019.

[5] NeelamMukhtar, Mohammad AbidKhan, and NadiaChiragh, "Lexicon-based approach

outperforms Supervised Machine Learning approach for Urdu Sentiment Analysis in
multiple domains", Telematics and Informatics, vol. 35, no. 8, pp. 2173-2183, December
2018.

[6] Y. Q. Lim, C. M. Lim, K. H. Gan and N. H. Samsudin, "Text Sentiment Analysis on Twitter
to Identify Positive or Negative Context in Addressing Inept Regulations on Social Media
Platform", 10th Symposium on Computer Applications and Industrial Electronics, 2020.
[7] N. H. Khun and H. A. Thant, "Visualization of Twitter Sentiment during the Period of US
Banned Huawei", 2019 International Conference on Advanced Information Technologies
ICAIT, 2019.
[8] A. Poornima and K. S. Priya, "A Comparative Sentiment Analysis of Sentence Embedding
Using Machine Learning Techniques", 6th ICACCS, 2020.