A novel asymmetric semantic similarity measurement for semantic job matching

摘要

Applying semantic similarity techniques in semantic matching applications can help to match information not only lexically but also semantically. In this paper, we extend the conventional semantic similarity measures for retrieving and ranking employment candidates in the case of semantic job matching. A framework for calculating asymmetric conceptual skill similarity is proposed, and validated in a use case of programming job matching. Within this case, a specific skills taxonomy has been formalized in Simple Knowledge Organization System (SKOS). A novel asymmetric semantic similarity measurement based on weighted-path-counting is proposed and validated in the use case. The proposed algorithms are evaluated by comparing them to user ranks, and our experimental results show that the proposed algorithms have better performance in ranking comparing to the conventional algorithms.