In processing of natural languages, most prominent branch of study is sentiment analysis. For finding the intension of text’s author, classification of text is performed in this analysis and intention may be of negative or positive type (criticism or admiration). In this work, four classification algorithms, namely, random forest (RF), K nearest neighbour (KNN), Support Vector Machine (SVM) and Naive Bayes (NB) are applied and their performances are compared. Sentimental reviews with negative or positive review are classified using these algorithms. Analysed the results comparison of classification methods like random forest (RF), K nearest neighbour (KNN), Support Vector Machine (SVM) and Naive Bayes (NB). From Amazon, dataset having 11,754 review sentences with labelling and 1.1M labelled review sentences with weak labelling are collected to forma dataset. Effectiveness of proposed method is exhibited in the results of experimentation.
Volume 12 | 03-Special Issue