Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 11 (2022) > Article

DOI: 10.14704/nq.2022.20.11.NQ66266

Optimal Feature Selection for Improving User Interest Identification in social media

R. Umamaheswari, M. Soranamageswari

Abstract

Interest based recommendation systems have become widely employed in practice as the number of Internet users and social network applications has increased in recent years. Considering a huge amount of data from LinkedIn and Twitter with growing number of users, it was of great significance to develop a real-time framework to recommend and monitor relevant tweets with respect to the interest of user. The interest for social network users was mined using association rules. A large number association rules mined from social networks was mainly depends on given coverage criteria. In association rule mining, a huge non-frequent interest terms were pruned only after identifying patterns of frequent and non-frequent patterns. In order to reduce the complexity of association rule mining process, the more relevant terms only selected by proposing Hybridized Competitive Swarm Optimizer and Gravitational Search Algorithm (CSO-GSA) in this paper.The optimal features are effectively retrieved by the CSO-GSA and then association rules are utilised for identifying the social network user’s interest.The preprocessing methods like aacronym expansion and negation termreplacementis also proposed for acquiring effective user interest through feature selection scheme and association rule mining algorithms.In this research, numerous relevant rules for human interest is identified. The numerical outcome of the proposed strategy is compared with existing state-of-art techniques. The proposed CSO with GSA outperforms the existing techniques.

Keywords

User interest, web crawling, twitter, LinkedIn, optimization, feature selection, CSO and GSA

Full Text

PDF

References

?>