Automatic Identification of Tag Types in a Resource-Based Learning Scenario

摘要

When users use tags they often have a rich semantic structure in mind, which can not be fully explicated using existing tagging systems. However, a tagging system needs to be simple in order to be successful, otherwise it will not be accepted by users. In our ELWMS.KOM system for the support of self-regulated Resource-Based Learning users can assign specific semantic types to the tags they use in order to manage their web-based learning resources. However studies have shown that most users would appreciate an automatic identification of tag types. In this paper we present a knowledge-based approach for the automatic identification of the tag types used in the ELWMS.KOM system. Evaluations conducted on different corpora show that the algorithm works with an overall accuracy of up to 84%.