Sentiment Analysis on Disputed Territory Discrepancies Using Machine Learning-based Text Mining Approach

Aug 3, 2026

Upload by Finn

Conducting real-time sentiment analysis faces practical barriers because accents and expressions vary continuously and direct quantification is difficult. This work evaluates supervised machine learning classifiers, including SVM, K-SVM, and Multinomial Naïve Bayes, for sentiment detection in tweets and news blogs about disputed territories. Models are trained on a Kaggle dataset and validated on real-time data. Kernel SVM yields stronger real-time performance with a higher share of positive sentiments. The approach supports correlating sentiment across countries to support early conflict-related predictions and reduce potential casualties.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Abstract

Keywords

Introduction

Sentiment analysis concepts and applications

Approaches: keyword-based, rule-based, machine learning, deep learning

Real-time challenges in sentiment recognition

VFAST Transactions on Software Engineering [http://vfast.org/journals/index.php/VTSE@ 2023](http://vfast.org/journals/index.php/VTSE@ 2023), ISSN(e): 2309-3978, ISSN(p): 2411-6246 Volume 11, Number 2, April-June 2023 pp:17-25 Sentiment Analysis on Disputed Territory Discrepancies Using Machine Learning-based Text Mining Approach Mustajib-ur-Rehman 1, Maria Bashir2 1Department of Computer Science, NFC Institute of Engineering & Technology, Multan, Pakistan. 2Faculty of Science and Technology, Norwegian University of Life Sciences. *Corresponding author [email: mustajeeb.rehman@nfciet.edu.pk](email: mustajeeb.rehman@nfciet.edu.pk) ABSTRACT Conducting a comparative study of real-time sentiment analysis poses a significant challenge due to the continuous variation in individuals' accents and the difficulty in quantifying them. In computer science, specifically supervised machine learning classifiers, researchers face the obstacle of lacking direct observation. This study explores the performance of well-known supervised machine learning classifiers such as SVM, K-SVM, and MultinomialNaïve Bayes. We utilize a comprehensive corpus of real-time Twitter tweets and news blogs related to disputed territories. These classifiers are trained on the Kaggle dataset for real-time sentiment analysis to achieve the highest accuracy and subsequently tested on real-time data. Notably, the kernel support vector machine performs better in real-time data, as evidenced by the substantial proportion ofpositive sentiments detected. Furthermore, our study ofers a pathway for young scholars to assess the real-time sentiment correlation between corresponding countries, enabling predictions ofpotential conflicts that may lead to significant casualties. KEYWORDS Text Classification, Sentiment Analysis, Supervised Machine Learning JOURNAL INFO HISTORY: Received: Received: April 10, 2023 Accepted: May 25, 2023 Published: May 30, 2023 1. INTRODUCTION Sentiment analysis is a technique used in natural language processing (NLP) to determine a text's emotional tone or attitude. It involves algorithms and statistical methods to analyze text data and identify the sentiment, which can be positive, negative, or neutral. Sentiment analysis is an essential tool used across various applications, such as social media monitoring, brand reputation management, market research, and customer feedback analysis. By analyzing the sentiment of customer reviews or reactions to news stories, businesses can recognize areas where they can improve their products or services and respond promptly to any negative feedback received. As a result of this analytical process employed during political campaigns or market research, significant insights into public opinions can be inferred using techniques like rule-based systems, machine learning algorithms, or deep learning models. These approaches comprehensively analyze various text features by examining the language, sentence structure, and contextual factors. Overall, sentiment analysis is a powerful tool that can help businesses and organizations to gain valuable findings from customer opinions and attitudes. Some most commonly used social media platforms for sentiment analysis are Twitter and Facebook because they are rich in text data. Machine Learning, NLP, and deep learning techniques can perform sentiment analysis on social media data in several ways. • Keyword-based analysis involves identifying specific keywords or phrases associated with positive or negative sentiment and then using these k • eywords to classify the sentiment of social media posts. For example, "love" may be associated with positive sentiment, while "hate" may be associated with negative sentiment. • Rule-based analysis involves defining rules or patterns to identify social media posts' sentiments. For example, a rule-based system may classify any post containing words like "happy" or "excited" as positive and any post containing

FAQs

What supervised classifiers are evaluated in the study?

The study evaluates SVM, K-SVM, and Multinomial Naïve Bayes for sentiment detection on real-time text data.

How is the model training and testing performed?

Classifiers are trained on a Kaggle dataset for real-time sentiment analysis and then tested on real-time data to assess performance.

Which classifier performs best on real-time data and what indicates this?

Kernel SVM performs better on real-time data, evidenced by a substantial proportion of positive sentiments detected.

Aug 3, 2026

Upload by Finn