Sentiment analysis is an approach for classifying the polarity of text source that is interested. Due to the high rate of using internet today, online user reviews which gradually increase in volume become an important data source for sentiment analysis studies in terms of accessibility and diversity. In this study, firstly, Multi-Layer Perceptron (MLP) algorithm is applied on online user reviews of an online bookstore for sentiment analysis using Python programming language. Afterwards, Naïve Bayes (NB), Support Vector Machines (SVM), and Logistic Regression (LR) algorithms are applied on the same dataset using RapidMiner data science software. Algorithms’ successes in classifying reviews are compared and Multi-Layer Perceptron becomes the algorithm showing the best results on this dataset.
Journal Type : Uluslararası
Relevant Articles | Author | # |
---|
Article | Author | # |
---|