Indonesian music fans group identification using social network analysis in Kaskus forum

The current practices of collecting information about market insight in music industry are through the sampling data collection using surveys, questionnaires and interviews. With the availibility of large-scale online conversation data, we can get faster; cheaper and more accurate data represent the actual phenomena of the market. The aim of this paper is to propose a new approach for helping the music industry parties to utilize network data mining methodology to identify one important insight in understanding music industry market, which is identification of music fans groups or conversations topics. Kaskus, the biggest online forum in Indonesia is a common media for millions Indonesian to discuss about many things, including their music preference. Last.fm is a social network of music enthusiasts. We mine Last.fm conversational topics in Kaskus and model those data using Social Network Analysis methodology based on graph model. We implement community detection algorithms to identify group identification. We show that network interaction pattern among the users in online forum can be used to help music industry parties identify the dominant topics and/or dominant music fans group in Indonesia.