Sentiment classification of customer reviews based on fuzzy logic

Abstract

Nowadays, e-commerce is growing fast, so product reviews have grown rapidly on the web. The large number of reviews makes it difficult for manufacturers or businesses to automatically classify them into different semantic orientations (positive, negative, and neutral). Most existing method utilize a list of opinion words for sentiment classification. whereas, this paper propose a fuzzy logic model to perform semantic classifications of customers review into the following sub-classes: very weak, weak, moderate, very strong and strong by combinations adjective, adverb and verb to increase holistic the accuracy of lexicon approach. Fuzzy logic, unlike statistical data mining techniques, not only allows using nonnumerical values also introduces the notion of linguistic variables. Using linguistic terms and variables will result in a more human oriented querying process.

A hybrid approach to the sentiment analysis problem at the sentence level

“…In addition to the most common rules, a number of authors, among them [41,59,69], have pointed out the fact that having rules for negation handling and to deal with the use of specific PoS particles, like 'but, despite, unless, ...' could positively affect the final outcome of a classification exercise.…”

A hybrid approach to sentiment analysis

“…A number of authors, among them [17,18,19], have pointed out to the fact that negation and the use of specific part-of-speech particles, like but, despite, unless, and others, and composition of functions -i.e. compose(v1, v2)-could affect the final outcome of a classification exercise.…”

A Hybrid Approach to Sentiment Analysis with Benchmarking Results

“…Indeed, a number of authors, among them [12,19,21], have pointed out the fact that negation and the use of specific part-of-speech particles, like 'but, despite, unless, ...' could affect the final outcome of a classification exercise. Thus, some rule strategies are needed to be put in place as the order of the different part-of-speech play a role in the semantic of a sentence.…”