Ontology-Based Social Media Talks Topic Classification (Twitter Case)

Abstract

In the era of digital communication, the use of Twitter as a customer service has been widely encountered. Companies have started to develop strategies around effective use of Twitter, one of which was to identify problems that customers frequently complain about. Twitter, with its straightforward tweet characteristics, will certainly contain sentences with very specific and easily recognizable keywords. These characteristics can be used as a basis for classifying tweets into certain topics. With a help of ontology, classification with keywords can be done automatically. The purpose of this paper is to design an ontology used as a basis for classifying tweets into certain topics related to the 4G telecommunications network in Indonesia and to evaluate performance of proposed classifier model.