• Version
• Download 3
• File Size 384.47 KB
• File Count 1
• Create Date 23 January 2026
• Last Updated 23 January 2026

Machine Learning-Based Sentiment Analysis of Online Hotel Reviews System

D. NANDHINI, MCA.,

(Assistant Professor, Master of Computer Applications)

A. ABDUL AASHIQ, MCA

Christ College of Engineering and Technology

Moolakulam, Oulgaret Municipality, Puducherry – 605010.

Abstract

The rapid growth of online hotel booking platforms has resulted in a massive increase in user-generated hotel reviews, making manual analysis of customer opinions inefficient and unreliable [1]. This paper proposes a Machine Learning-Based Sentiment Analysis System for Online Hotel Reviews that automatically classifies customer feedback into positive, negative, or neutral sentiments using natural language processing and supervised machine learning techniques [2]. The system performs text preprocessing operations including tokenization, stop-word removal, normalization, and TF-IDF vectorization to transform unstructured review text into numerical feature representations [3]. These features are used to train machine learning models such as Support Vector Machine and Naive Bayes, where the SVM model achieves an accuracy of approximately 97% on the hotel review dataset [4][5]. In addition, the system supports fake review identification and aspect-based sentiment analysis focusing on key hotel attributes such as service, cleanliness, and location [6]. The proposed system is implemented as a web-based application using Python, Flask, Scikit-learn, NLTK, and SQLite [7][8], providing users with real-time sentiment analysis, confidence scores, and result visualization. Experimental results demonstrate that TF-IDF-based feature extraction combined with SVM offers reliable and balanced sentiment classification, making the system a practical and scalable solution for opinion mining and decision support in the hospitality industry [1][4].

Keywords

Sentiment analysis, online hotel reviews, machine learning, support vector machine, TF-IDF, natural language processing, fake review detection, web-based application.