Toward Automatic Semantic Annotating and Pattern Mining for Domain Knowledge Acquisition

Due to the high complexity of natural language, acquisition of high quality knowledge for the purpose of fine-grained data processing still mainly relies on manual labour at present, which is extremely laborious and time consuming. In this paper, a new automatic approach using semantic annotating and pattern mining is proposed to assist engineers for domain knowledge acquisition. This approach uses Minipar to label sentences processed from domain texts. Based on the dependency relations, structural patterns are extracted and semantic bank is applied to annotate and represent concepts with semantic labels considering sentence contexts. The approach further learns and assigns relations to previously extracted concepts by pattern matching. The involved concepts and semantic labels with learned relations together, as extracted knowledge, enrich domain knowledge base. Preliminary experiments on Yahoo! Data in “heart diseases” category showed that the proposed approach is feasible for automatic domain knowledge acquisition.