SENTIMENT ANALYSIS USING HYBRID MACHINE LEARNING APPROACH

SENTIMENT ANALYSIS USING HYBRID MACHINE LEARNING APPROACH

Soham Dalvi1, Suyash Gharge2, Prasad Gorhe3, Saurabh Dev4
Prof. S.P. Jadhav5
Department Of Information Technology, Sinhgad College Of Engineering, Pune, India
---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Sentiment Analysis (SA) is a rapidly evolving research field that faces challenges in keeping pace with its growth. SA explores how individuals express opinions and attitudes toward various subjects, events, and products through textual evaluations on platforms like blogs and forums. Amazon, a leading online retailer, allows users to openly rate products and submit reviews, vital for consumer decision-making. Analyzing these reviews for sentiment, positive or negative, aids decision-making from purchases to film reviews. Our project employs a hybrid machine learning approach to refine accuracy in sentiment analysis. Our project endeavors to categorize reviews according to their sentiments. Leveraging diverse machine learning models, our goal is to accurately classify the provided reviews into distinct groups representing positive, negative, or neutral sentiments. Moreover, our endeavor extends beyond mere classification; we also intend to
conduct a comparative analysis of various machine learning models to assess their performance and accuracy in sentiment analysis.

Key words: Sentiment Analysis, Sentiment Classification, Opinion Mining, Machine Learning, Polarity Classification, Supervised Algorithms, Amazon Classification, Ensemble Learning, Hybrid Machine Learning.