A generic lexicon-based framework for sentiment analysis. Issue 5 (3rd July 2020)

• Record Type: Journal Article
• Authors: Moussa, Mohammed Elsaid; Mohamed, Ensaf Hussein; Haggag, Mohamed Hassan
• Is Part Of: International journal of computers and applications. Volume 42:Issue 5(2020)
• Page Start: 463
• Page End: 473
• Publication Date: 2020-07-03
• Subjects: Natural language processing -- sentiment analysis -- lexicons -- lexicon-based framework; Computers -- Periodicals; Computer software -- Periodicals; Computer networks -- Periodicals; Multimedia systems -- Periodicals; Internet -- Periodicals; World Wide Web -- Periodicals; Minicomputers -- Periodicals; Microcomputers -- Periodicals; 004.05
• DOI: 10.1080/1206212X.2018.1483813
• Languages: English
• ISSNs: 1206-212X

Abstract

ABSTRACT: On the internet, people share their opinions and sentiments in many forms of products or services. This behavior creates an extensive collection of opinions and views in the form of texts, which needs to be analyzed to know the efficacy of the product or service. Traditional lexicon-based approaches suffer from scalability and are limited by unreliable sentiment lexicons that are commonly created manually by experts. So, we proposed a generic, extendable, domain-independent lexicon-based framework for sentiment analysis that overcomes scalability, limitedness, and beats existing models in accuracy. The framework engine sums up the sentiment intensities based on a list of different lexicons, and a dedicated neutral bias module handles the misinterpreted neutral polarities. Extensive experiments on five different lexicons and using a sample sentiment engine show significant improvements in accuracy over traditional lexicon-based sentiment approaches.