Scribd
Upload
3 views11 pages

OTCON4 Template

The document presents a study on sentiment analysis of political tweets using machine learning techniques, specifically Naive Bayes, SVM, and Decision Tree algorithms. The study found that Naive Bayes achieved the highest accuracy of 85.12%, demonstrating the effectiveness of machine learning in analyzing public political opinions. The research highlights the importance of automated sentiment detection in understanding political discourse on social media.

Uploaded by

Honey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views11 pages

OTCON4 Template

The document presents a study on sentiment analysis of political tweets using machine learning techniques, specifically Naive Bayes, SVM, and Decision Tree algorithms. The study found that Naive Bayes achieved the highest accuracy of 85.12%, demonstrating the effectiveness of machine learning in analyzing public political opinions. The research highlights the importance of automated sentiment detection in understanding political discourse on social media.

Uploaded by

Honey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 11

2025 4th OPJU International Technology Conference (OTCON)

on Smart Computing for Innovation and Advancement in Industry 5.0


(9 - 11 April, 2025)

Analysis of Sentiment in Political


Tweets on Social Networks using
Machine Learning Approaches

BY-
Vivek Priyadarshi Dash
Manjushree Nayak
Salim Amirali Jiwani
Umashankar Ghugar
Rakesh Nayak

OP Jindal University, Raigarh, India 1


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Abstract

• Twitter is a major platform for expressing political


opinions.
• Sentiment analysis classifies tweets as positive,
negative, or neutral.
• Machine learning techniques were applied to classify
sentiments.
• Algorithms: Naive Bayes, SVM, Decision Tree.
• Naive Bayes achieved highest accuracy in this analysis.
OP Jindal University, Raigarh, India 2
2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Introduction

• Social media plays a vital role in political discourse.


• Analyzing sentiments helps understand public political opinion.
• Sentiment analysis detects emotions from textual data.
• Political tweets are often opinion-rich and timely.
• ML enables automated and scalable sentiment detection.

OP Jindal University, Raigarh, India 3


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Literature Review

• Previous research used hashtags and tweet polarity to


detect sentiments.
• Techniques like SVM, lexicon methods, and hybrid models
were applied.
• Studies showed political trends can be predicted through SA.
• Challenges like sarcasm and short length of tweets are noted.
• Existing systems proved effective, but can be further enhanced.

OP Jindal University, Raigarh, India 4


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Problem Identification
• Political tweets are increasing and influence public
perception.
• Manual monitoring of sentiments is not feasible.
• Political tweets are often ambiguous or emotionally
charged.
• There's a need to accurately classify tweet sentiment.
• Goal: Build a system that analyzes sentiments using
ML.

OP Jindal University, Raigarh, India 5


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Methodology

•Preprocessing: Stop word removal, stemming, tokenization.


•Feature extraction: TF-IDF vectorization of tweets.
•Models used: Naive Bayes, SVM, Decision Tree.
•Implementation tools: Python, scikit-learn, NLTK.
•Training/testing split ensures fair model evaluation

OP Jindal University, Raigarh, India 6


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Dataset

•13,000 political tweets collected using Twitter API.


•Tweets labeled as positive, negative, or neutral.
•Text cleaning included noise and punctuation removal.
•Preprocessing converted tweets into numeric vectors.
•Dataset ensured class balance and relevancy to politics.

OP Jindal University, Raigarh, India 7


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Results and Analysis

• Naive Bayes: 85.12% accuracy (best performing).


• SVM: 82.64% accuracy, slightly lower than NB.
• Decision Tree: 78.26% accuracy (least effective).
• NB performed better due to its efficiency with
sparse data.
• Results validated the effectiveness of ML in SA
tasks.

OP Jindal University, Raigarh, India 8


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Tables
Machine Accuracy(% Precision Recall F1-Score
Learning )
Model
Native Bayes 85.12 0.86 0.84 0.85

SVM 82.64 0.83 0.81 0.82

Decision Tree 78.26 0.79 0.77 0.78

OP Jindal University, Raigarh, India 9


2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

Conclusion

• ML models can effectively classify political tweet


sentiments.
• Naive Bayes proved to be the most accurate and
reliable.
• Sentiment analysis aids in gauging public political
mood.
• The study lays foundation for larger-scale political
analytics.
• Future work: Deep learning, sarcasm detection,
multilingual tweets. OP Jindal University, Raigarh, India 10
2025 4th OPJU International Technology Conference (OTCON)
on Smart Computing for Innovation and Advancement in Industry 5.0
(9 - 11 April, 2025)

References

•[1] Research articles on SA and political ML models.


•[2] Python libraries used: scikit-learn, NLTK.
•[3] Datasets obtained through Twitter API.
•[4] Prior SA studies on political social media trends.
•[5] Books and journal articles cited in the paper.

OP Jindal University, Raigarh, India 11

You might also like