Evaluating the importance of Web comments through metrics extraction and opinion mining

Abstract

The evolution of e-commerce and Online Social Networks made significant g rowth of t he W eb a nd as consequence, available information increase quite every day, making the task of analyzing the reviews manually almost impossible for the decision-making process. Due to the amount of information, the creation of automatic methods of knowledge extraction and data mining has become necessary. This paper presents a Web application prototype where from a review are returned the feeling (positive, negative or neutral), its features and other analysis metrics using Natural Language Processing and Sentiment Analysis in order to define t he m ost important comments to be taken into consideration in the decision-making process. Experiments show efficacy i n t he p recision o f reviews with negative polarity and recall of reviews with positive polarity in 84.93% and 94.33% respectively and the most important comments were found in a measure considered satisfactory of 50% in F-Measure in both positive and neutral polarities.