Optimized Machine Learning framework for Sentiment Analysis for Amazon Product Reviews
Main Article Content
Abstract
Sentiment Analysis (SA) commonly referred to as opinion mining, is a subset of natural language processing that focuses on identifying views and attitudes inside a document. Currently, this paper examines the SA of the Amazon product reviews where Machine Learning (ML) algorithms will be employed to categorize the user sentiments as positive, negative, or neutral. The authors of the research use a dataset of Amazon product reviews for the text analysis to avoid featuring insignificant words, the text data is tokenized, stop-words are removed, and PoS. Here, KNN, SVM, RF are used for the development of models based on machine learning algorithms. Moreover, there are text preprocessing applied method such as the TF-IDF and Bag of Word. Metrics used to measure accuracy of the models are; ????????????????????????????????????, ????????????????????????????????????????,????????????????????????????, ????1????????????????????. The results of the KNN and SVM models were compared with the findings of the RF model. Using the data set, the complementary RF model achieved higher ???????????????????????????????????? of 98.83%, ???????????????????????????????????????? of 97.70%, ???????????????????????????? of 98.64%, and ????1???????????????????? of 98.17%.
Article Details

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.