OPINION MINING FOR SOCIAL NETWORKING SITE

Abstract

Social networking platforms generate large volumes of user opinions through posts, comments, reviews, and discussions, making manual analysis difficult, time-consuming, and inefficient. This paper presents an automated opinion mining system that integrates Natural Language Processing (NLP) and Machine Learning (ML) techniques to classify social media opinions into positive, negative, and neutral categories. The proposed framework performs data collection, text cleaning, tokenization, stop-word removal, stemming, and TF-IDF-based feature extraction to transform unstructured textual data into meaningful numerical representations. A Naive Bayes classifier is then employed for sentiment classification, while the results are stored and presented through a user-friendly interface and graphical reports. The system is implemented using Python, Flask, NLTK, Scikit-learn, and MySQL. Evaluation considers accuracy, precision, recall, and F1-score. The framework provides scalable and efficient opinion analysis for organizations, researchers, and decision-makers. Keywords: Opinion Mining, Sentiment Analysis, Natural Language Processing, Machine Learning, TF-IDF, Naive Bayes.