CDS: Collaborative distant supervision for Twitter account classification. (15th October 2017)

• Record Type: Journal Article
• Authors: Cui, Lishan; Zhang, Xiuzhen; Qin, A.K.; Sellis, Timos; Wu, Lifang
• Is Part Of: Expert systems with applications. Volume 83(2017)
• Page Start: 94
• Page End: 103
• Publication Date: 2017-10-15
• Subjects: Twitter -- Classification -- Distant supervision -- Active learning -- Semi-supervised learning; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2017.03.075
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Novel distant supervision-based approach to Twitter account classification. Collaborative learning of distant supervision, active and semi-supervised learning. Heuristics for automatically labelling Twitter accounts. Generic strategy identifying false positives and false negatives from automatic labelling. Abstract: Individuals use Twitter for personal communication, whereas businesses, politicians and celebrities use Twitter for branding purposes. Distinguishing Personal from Branding Twitter accounts is important for Twitter analytics. Existing studies of Twitter account classification apply classical supervised learning, which requires intensive manual annotation for training. In this paper, we propose CDS (Collaborative Distant Supervision), a novel learning scheme for Twitter account classification that does not require intensive manual labelling. Twitter accounts are automatically labelled using heuristics for distant supervision learning. To achieve effective learning from heuristic labels, active learning is applied to identify and correct false positive labels, and semi-supervised learning is applied to further use false negatives missed by labelling heuristics for learning. Extensive experiments on Twitter data showed that CDS achieved high classification accuracy.