Feature-based Approach for Sentiment Analysis of Social Networks

Abstract

In this chapter, the performance of the proposed approach is evaluated by conducting four different experiments.These experiments are designed to study the effect of each one of the key challenges individually and all together on the overall performance of the proposed approach.Hence, the performance of the proposed approach is evaluated by comparing the results obtained from feature-based sentimentanalysis before and after considering emoticons detection, negation handling, and spam reviews detection. The performanceof the proposed approach is measured using four different types of evaluation metrics which are accuracy,precision, recall, and f1 score. In addition to that, the effectiveness of the proposed approach is verified against three different datasets of different sizes. Moreover, a comparison between the performance ofour proposed spam reviews detection method and the performance of several state-of-the-art spam reviews detection methods is performed. This comparison showed that the performance of the proposed spam reviews detection method outperforms the performance of the state-of-the-art methods.Finally and according to the overall results achieved, it has been noticed that the overall performance of the proposed feature-based sentiment analysis approach has been improved after applying the feature extraction method “extracting frequent nouns by applying Apriori algorithm” and after considering the three main challenges negation handling, emoticons and spam reviews detection together.