Seminar Report

On

Twitter Sentiment Anaysis

By

Vrushank Chaphadkar

Under the Guidance of

Dr Prakash Kene

Masters in Computer Application

P. E.S Modern College of Engineering
Pune-411005
Savitribai Phule Pune University
A.Y.2022-23
 Progressive Education Society's
Modern College of Engineering,
Pune-5
Department of MCA

Certificate

This is to certify that VRUSHANK CHAPHADKAR of Master of Computer

Application have successfully completed the seminar work titled TWITTER
SENTIMENT ANALYSIS during the academic year 2022-23. This report is submitted
as partial fulfilment of the requirement of degree in MCA Engineering of Savitribai
Phule Pune University.

(Prof.Dr.Mrs.K. R.Joshi) (Dr.Mrs.P.A.Muley)

Principal Head of Department

(GUIDE NAME)
Seminar Guide
 Progressive Education Society's
Modern College of Engineering,
Pune-5
Department of MCA

ACKNOWLEDGEMENT

This seminar presentation is not an individual task. I express a deep sense of gratitude
to the Principal Prof. Dr. Mrs. K.R. Joshi, HOD MCA (Engg.) department, Prof. Dr.
Mrs. Pradnya Muley and My seminar Guide DR.PRAKASH KENE for their
support, encouragement and timely guidance.

VRUSHANK CHAPHADKAR
CLASS: SEM IV MCA
ROLL NO: 52110
 Table of contents

Chapte Title Page No.

r No.

1. Introduction 1

2. Literature Survey 2

3. Architecture/Components/ 4
Comparison/ Characteristics

4. Advantages 13

5. Disadvantages 15

6. Applications 17

7. Future Enhancement 19

8. Conclusion 20

9. References 20
 CHAPTER 1 INTRODUCTION

Twitter, a popular microblogging platform, has revolutionized the way people express
their opinions and sentiments on various topics in real-time. With millions of users
actively engaging on Twitter, it has become a valuable source of data for sentiment
analysis. Sentiment analysis, also known as opinion mining, is the process of extracting
subjective information from text and determining the emotional tone behind it.

Understanding public sentiment on Twitter has numerous applications, ranging from

brand reputation management and market research to political analysis and public
opinion tracking. By analyzing tweets, researchers, businesses, and policymakers can
gain insights into people's attitudes, emotions, and opinions towards specific subjects,
products, events, or individuals.

The goal of Twitter sentiment analysis is to automatically classify tweets into positive,
negative, or neutral categories based on the expressed sentiment. This classification can
be performed using various techniques, including natural language processing (NLP),
machine learning, and deep learning algorithms. These methods enable the analysis of
large volumes of tweets efficiently and accurately, providing valuable insights into
public sentiment trends.

However, analyzing sentiment on Twitter poses several challenges. Firstly, tweets are
often short and informal, containing abbreviations, slang, emojis, and other
unconventional language usage. Secondly, Twitter data is characterized by noise,
including spam, irrelevant content, and sarcastic or ironic expressions. Additionally,
tweets often contain contextual information, such as mentions, hashtags, and retweets,
which can influence sentiment analysis.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 1| Page

 CHAPTER 2 LITERATURE SURVEY

Title: "Sentiment Analysis of Twitter Data for Predicting Stock Market Movements"
(2011) Authors: Johan Bollen, Huina Mao, Xiao-Jun Zeng
Summary: This paper focuses on the correlation between Twitter sentiment and stock
market movements. The authors propose a method to analyze sentiment in Twitter data
and demonstrate its potential for predicting stock market trends.

Title: "Twitter Mood Predicts the Stock Market" (2011)

Authors: Johan Bollen, Huina Mao, Xiao-Jun Zeng
Summary: This study explores the relationship between public mood expressed on
Twitter and the stock market. The authors propose a method to extract mood from
Twitter data and demonstrate its effectiveness in predicting stock market trends.

Title: "Twitter Sentiment Analysis: The Good the Bad and the OMG!" (2013)
Authors: Efthymios Kouloumpis, Theresa Wilson, Johanna Moore
Summary: This paper presents a comprehensive analysis of sentiment analysis
techniques applied to Twitter data. The authors evaluate various methods and discuss the
challenges and limitations of sentiment analysis in the context of Twitter.

Title: "Sentiment Analysis and Opinion Mining on Twitter: A Survey" (2014)

Authors: Ahmad Ali, Asim Karim, Shah Nawaz
Summary: This survey paper provides an overview of sentiment analysis and opinion
mining techniques specifically focused on Twitter data. The authors review different
approaches, datasets, and evaluation metrics used in sentiment analysis research on
Twitter.

Title: "Twitter Sentiment Analysis: A Hybrid Approach Using NLP Techniques" (2015)

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 2| Page

Authors: Fatima Zohra Chelghoum, Fadila Bentayeb, Omar Boussaid
Summary: This research proposes a hybrid approach combining Natural Language
Processing (NLP) techniques for sentiment analysis on Twitter data. The authors explore
the effectiveness of combining linguistic and machine learning features for improved
sentiment classification.

Title: "Twitter Sentiment Analysis: A Comparative Study" (2016)

Authors: Muhammad Kamran Malik, Khurum Nazir Junejo
Summary: This comparative study analyzes the performance of various machine learning
algorithms for sentiment analysis on Twitter data. The authors compare the accuracy,
precision, recall, and F1-score of different approaches and discuss their findings.

Title: "Sentiment Analysis of Twitter Data: A Review" (2017)

Authors: Shreya Kumari, Devansh Arpit, Hrishikesh Barua
Summary: This review paper provides an overview of sentiment analysis techniques
applied to Twitter data. The authors discuss the challenges, methodologies, and
applications of sentiment analysis in the context of Twitter and identify future research
directions.

Title: "Aspect-Level Sentiment Analysis on Twitter Data" (2018)

Authors: Soujanya Poria, Navonil Majumder, Devamanyu Hazarika, et al.
Summary: Paper focuses on aspect-level sentiment analysis on Twitter data, which aims
to determine sentiment polarity for specific aspects or entities mentioned in tweets. The
authors propose a novel model and dataset for aspect-level sentiment analysis.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 3| Page

 CHAPTER 3 ARCHITECTURE/IMPLEMENTATION

Twitter sentiment analysis typically involves analyzing the sentiment or opinion

expressed in tweets. The architecture for Twitter sentiment analysis can vary
depending on the specific requirements and the complexity of the task. Here is a
high-level overview of a common architecture used for Twitter sentiment analysis:

Data Collection: The first step is to collect a large dataset of tweets. This can be done
by using the Twitter API or by using publicly available datasets.

1.Preprocessing: The collected tweets need to be preprocessed before analysis. This

step involves removing irrelevant information such as URLs, special characters,
hashtags, and mentions. It may also include tokenization (splitting text into
individual words or tokens), removing stop words (common words like "and," "the,"
etc.), and normalizing the text (converting uppercase letters to lowercase, removing
punctuation marks, etc.).

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 4| Page

2.Feature Extraction: After preprocessing, relevant features need to be extracted
from the tweets. This step involves representing the text data in a numerical form
that machine learning algorithms can process. Some common feature extraction
techniques for sentiment analysis include bag-of-words (representing each tweet as a
vector of word frequencies), n-grams (capturing sequences of words), and word
embeddings (representing words as dense vectors in a continuous space).

3.Sentiment Classification: Once the features are extracted, a classification

algorithm is used to classify the sentiment of each tweet. Various machine learning
algorithms can be employed for this task, such as Support Vector Machines (SVM),
Naive Bayes, Random Forests, or deep learning models like Recurrent Neural
Networks (RNNs) or Convolutional Neural Networks (CNNs).

4.Model Training: The sentiment classification model is trained using a labeled

dataset where each tweet is associated with a sentiment label (positive, negative,
neutral). The training data is split into a training set and a validation set to evaluate
the model's performance and tune hyperparameters.

5.Model Evaluation: The trained model is evaluated on a separate test dataset to

assess its performance. Common evaluation metrics for sentiment analysis include
accuracy, precision, recall, F1 score, and confusion matrix.

6. Deployment: Once the model is trained and evaluated, it can be deployed to

analyze real-time or incoming tweets. The deployed model takes a tweet as input,
applies the same preprocessing and feature extraction steps, and predicts the
sentiment label based on the trained classification algorithm.

It's important to note that this is a generalized architecture, and specific

implementations may vary based on the requirements and constraints of the

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 5| Page

sentiment analysis task. Additionally, advancements in deep learning and
transformer-based architectures, such as BERT or GPT, have shown promising
results in sentiment analysis tasks, including those related to Twitter data.

Flow Chart Diagram

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 6| Page

Screen Visuals:

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 7| Page

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 8| Page
PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 9| Page
UserManual:
Step1 –
Enter Twitter search

Step2 –
Authorize the twitter API consumer key connection

Step3 –

Load packages functions and global variables for processing the data
Step4 –

Now the model extract the relevant data from the tweets fetched

Step5 –
Function [def clean()] cleans the tweets and removes unnecessary symbols,

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 10| Page

 brackets, alpha-numerics and converts the tweets into small case.

Step6 -
Keywords are extracted using TF-IDF (Term frequency- Inverse Document
Frequency)
Step7-
Naive bayes and SVM algorithm is applied for classification of tweets
Step8 –
Naive bayes algorithm uses single word classification technique
Step9 –
Support vector machine plots the measures of the plot.

Step10 –
Machine learning model predicts the result of tweet whether it is positive ,
negative or neutral

Step 11-
Display the result
Step11 –

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 11| Page

 Start the second module which is “Crime Check”.

Step 12-

Input the crime keyword

Step 13-
Select the location to check the crime

Step 14-
The model displays tweets, username, location of the tweet handler.

Step 15-
We get the polarity and subjectivity of the tweet which determines the fact and
opinion about the tweet.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 12| Page

CHAPTER 4 ADVANTAGES

Sentiment analysis is a technology that automatically detects emotions in social media

interactions. Companies can use sentiment analysis to know their customers, stay on
top of what is being said about their business – and their rival – and find new trends in
the market by paying careful attention to the voice of their customer on twitter.

There are currently over 206 million marketable daily Twitter users globally. With
such a large potential marketplace for their goods and services, businesses will
undoubtedly engage substantially in studying the behavioral patterns of their core
demographics in order to personalize their offers to generate profit and a loyal client
base.

1. Find out your brand perception: The Twitter sentiment analysis can provide an
overview of your brand’s opinion. You can find out what people are talking
about your company and its customers. Understanding brand perception can
help you spot potential difficulties as well as capitalize on hidden possibilities.
A single mention from a well-known blogger can sometimes propel your brand
to prominence and give your business a boost.

2. Build stronger customer relationships: One of the most important advantages of

opinion mining is that it gives firms more insight into client behavior.
You can comprehend how your audience is feeling at any given time and
contextualize the results – what made them pleased, thrilled, furious,
dissatisfied, or indifferent? Did you make a mistake? Or, even better, perhaps
you performed an outstanding job with your most recent campaign and people
can’t get enough of it.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 13| Page

 3. Offer better customer service: Delivering a consistently excellent customer
experience is a demanding undertaking, and it becomes even more difficult
when your company grows, because it may be impossible to maintain the same
level of quality at scale while keeping every customer delighted. As marketers,
we frequently employ fantastic tools like buyer personas, user journey charts,
and categorization to aggregate and better comprehend the profiles of our
typical consumers.

4. Identify key emotional triggers: Human decisions are influenced by emotional

stimuli. You can use sentiment analysis to determine which communications
and discussions work as emotional triggers, changing client mood. For example,
the phrase ‘Please wait’ frequently causes consumer displeasure. Perhaps
utilizing emojis improves the general tone of the discourse. Understanding what
messages elicit certain sentiments in your clients can assist you provide better
service and creating effective marketing resources.

6. Manage crisis better: One of the advantages of sentiment analysis is that it

allows you to track your customers’ replies or views about a brand on a regular
basis. Among the several goals of sentiment analysis, one is to prevent growing
complaints, which enables effective and quick crisis management. Timely
preventive steps are critical because they aid in the eradication of online
communication crises, which can quickly spread across the Internet. Since
sentiment analysis enables firms to keep a close check on any negative threads
or comments posted online, any issues or crises can be addressed before they
escalate.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 14| Page

 CHAPTER 5 DISADVANTAGES

1.Sarcasm Detection: In sarcastic text, people express their negative sentiments using
positive words. This fact allows sarcasm to easily cheat sentiment analysis models
unless they’re specifically designed to take its possibility into account. Sarcasm occurs
most often in user-generated content such as Facebook comments, tweets, etc.
Sarcasm detection in sentiment analysis is very difficult to accomplish without having
a good understanding of the context of the situation, the specific topic, and the
environment.It can be hard to understand not only for a machine but also for a human.
The continuous variation in the words used in sarcastic sentences makes it hard to
successfully train sentiment analysis models. Common topics, interests, and historical
information must be shared between two people to make sarcasm available.

2.Negation Detection: In linguistics, negation is a way of reversing the polarity of

words, phrases, and even sentences. Researchers use different linguistic rules to
identify whether negation is occurring, but it’s also important to determine the range
of the words that are affected by negation words. There is no fixed size for the scope
of affected words. For example, in the sentence “The show was not interesting,” the
scope is only the next word after the negation word. But for sentences like “I do not
call this film a comedy movie,” the effect of the negation word “not” is until the end
of the sentence. The original meaning of the words changes if a positive or negative
word falls inside the scope of negation—in that case, opposite polarity will be
returned. The simplest approach for dealing with negation in a sentence, which is used
in most state-of-the-art sentiment analysis techniques, is marking as negated all the
words from a negation cue to the next punctuation token. The effectiveness of the
negation model can be changed because of the specific construction of language in
different contexts.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 15| Page

3.Word Ambiguity: Word ambiguity is another pitfall you’ll face working on a
sentiment analysis problem. The problem of word ambiguity is the impossibility to
define polarity in advance because the polarity for some words is strongly dependent
on the sentence context.Lexicon-based sentiment analysis approaches are popular
among existing methods. An opinion lexicon contains opinion words with their
polarity value. There are some public opinion lexicons available on the internet:
SentiWordNet, General Inquirer, and SenticNet, among others. Because word polarity
varies in different domains, it is impossible to develop a universal opinion lexicon that
has a polarity for every word. For example:
1. “The story is unpredictable.”
2. “The steering wheel is unpredictable.”
These two examples show how context affects opinion word sentiment. In the
first example, the word polarity of “unpredictable” is predicted as positive. In
the second, the same word’s polarity is negative.

4.Multipolarity: Sometimes, a given sentence or document—or whatever unit of text

we would like to analyze—will exhibit multipolarity. In these cases, having only the
total result of the analysis can be misleading, very much like how an average can
sometimes hide valuable information about all the numbers that went into it.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 16| Page

 CHAPTER 6 APPLICATIONS

1.Brand Experience Insight: You can gather valuable brand experience insights that
can give you a peep into hidden market sentiment about your brand and what
customers expect from you. These insights are important because they allow you to
understand market-gap issues, retain customers, build a loyal customer base, and
increase sales conversions.

2.Improve Customer Service: Benefits of sentiment analysis include improving

customer service through customer experience analysis. Emotion mining chatbot
histories, customer service call transcripts, customer complaint emails, returns and
refunds customer comments, and customer surveys, all can be harnessed to improve
customer satisfaction and help you develop brilliant customer experience.

3.News Trend Analysis: Transaction tracking systems can enhance tax administration
by tracking financial transactions and identifying potential tax evasion or
underreporting. It facilitates accurate tax collection and minimizes the scope for tax
fraud.

4.Public Infrastructure Projects: Sentiment analysis is used to extract emerging trends

from news websites, videos, articles and magazines, online platforms that include
blogs, social media like Twitter and Facebook, and other sources to anticipate market
behavior. This includes news on current affairs such as new political scenarios,
international crude-oil trading, and share movements of enterprises.

5.Real-Time Sentiment Insights: Sentiment analysis benefits also extend to real-time

analysis of sentiment from live video streams and live texts and comments. This is
used to analyze audience participation and satisfaction during live shows, live

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 17| Page

corporate events, seminars, promotional events like trade and car shows, live radio
broadcasts, and others.

6.Education and Scholarship Programs: Transaction tracking systems can be employed

to track financial disbursements for education and scholarship programs. It ensures
that funds reach deserving students, monitors their utilization, and reduces the chances
of misappropriation or misuse.

7.Product Development: Emotion mining from customer feedback data, surveys, news
reports and articles, social media listening, and other sources can give you clever
insights into how you can improve your product so that it reaches more audiences.
This is also very important when launching a new product, opening a store at a new
location, changing business models, and such.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 18| Page

 CHAPTER 7 FUTURE ENHANCEMENT

1.Identifying and Predicting Market Trends: It enables you to analyze large amounts
of market research data in order to spot emerging trends and better understand
consumer buying habits. This type of practice can help you navigate the complicated
world of stock market trading and make decisions based on market sentiment.

2.Keeping an eye on the brand’s image: Sentiment analysis is frequently used to

investigate user perceptions of a product or topic. You can also use it to conduct a
product analysis and provide all relevant data to the development teams.

3. Examining public opinion polls and political polls :To predict the outcome of an
election, anyone can use sentiment analysis to compile and analyze large amounts of
text data, such as news, social media, opinions, and suggestions.

4. Data from customer feedback is being analyzed: Data from customer feedback can
be used to identify areas for improvement. Sentiment analysis can help you extract
value and insights from customer feedback data, as well as develop effective customer
satisfaction strategies.

5.Observing and analyzing conversations on social media: Conversations on social

media are a gold mine of information. Look at conversations about your brand on
social media to see what your customers are saying with sentiment analysis; this can
help any company plan its future strategies much more effectively.

6. Employee Turnover Reduction: Analyze large amounts of employee feedback data

to determine employee satisfaction levels. The insights are used by the sentiment
analysis tool to boost morale and productivity while also informing you of how
employees are feeling.

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 19| Page

CHAPTER 8 CONCLUSION

The Sentiment analysis is incredibly common technology in today’s world. Most of

works has been tired these fields. In all these paper several analysis has been done on
sentiment analysis of social media knowledge and that they have used several
techniques and technology like Naive Thomas Bayes, Hadoop Framework and Map
scale back programming, TFIDF, lexicon-based, SVM classifier, Levenshtein
algorithmic rule. So in my analysis work I need to extend the accuracy and potency
and reduce the time overwhelming to analysis of Posts, tweets on a twitter.

CHAPTER 9 REFERANCES

Bhuvan, U., & Apte, A. (2019). Blockchain for Government Services: A Systematic
Literature Review. International Journal of Computer Science and Information
Security, 17(3), 128-136.

Zyskind, G., Nathan, O., & Pentland, A. (2015). Decentralizing Privacy: Using
Blockchain to Protect Personal Data. In Security and Privacy Workshops (SPW), 2015
IEEE (pp. 180-184). IEEE.

Azaria, A., Ekblaw, A., Vieira, T., & Lippman, A. (2016). MedRec: Using Blockchain
for Medical Data Access and Permission Management. In 2016 2nd International
Conference on Open and Big Data (OBD) (pp. 25-30). IEEE.

Zheng, Z., Xie, S., Dai, H. N., Chen, X., & Wang, H. (2017). An Overview of
Blockchain Technology: Architecture, Consensus, and Future Trends. In 2017 IEEE
International Congress on Big Data (BigData Congress) (pp. 557-564). IEEE

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 20| Page

PES MODERN COLLEGE OF ENGINEERING , MCA DEPARTMENT 21| Page