AN APPROACH FOR DEVELOPING INTELLIGENT SYSTEMS FOR SENTIMENT ANALYSIS OVER SOCIAL NETWORKS

摘要

This work presents a novel approach towards buildingrnintelligent systems for detecting users’ perception onrncompanies brands through the use of social networks.rnUsers’ perception is crucial for detecting the possiblernsuccess or failure of a company in the web 2.0 era morernthan in the past. For this reason we propose here anrnapproach for building intelligent systems that are able torndetect users’ opinion about a brand exploring the Twitterrnsocial network. By using Twitter messages the proposedrnapproach employs an automatic technique based onrnsemantic differential together with natural languagernprocessing and graph ontology exploration for detectingrnusers’ perception of brands. We demonstrate how ourrnsystem could provide companies with knowledge aboutrnusers’ perception for decision making, and knowledgerndiscovery applications, by presenting two use cases.