Research on Intelligent Semantic Search Based on HowNet

ABSTRACT

HDWiki is similar to the famous Wikipedia but as being a major encyclopedia in Chinese, it includes vast investment of manual effort and judgment. This structured information is divided into three categories: concept, relationship and entity. This paper describes a novel symbol oriented search that facilitates the user to format unambiguous query, especially to format complex logical query which needs several queries to get answer on traditional search but once incorporated here, it builds a bridge between what the user knows and what the user wishes and between the intent of user query and the understanding of search engine. This facilitation and understanding based on the ternary relationships are everywhere in HDWiki and HowNet powered semantic similarity measure for knowledge matching. A logical semantic oriented search system is realized and a detailed user study is conducted with users and it is found that this logical semantic oriented search and its underlying knowledge base provide significant advantages over conventional search. It offers assistance to almost every query especially complex logical queries, making the query entry more clear and efficient for user as well as for search engine, broadening the processing scope of complex logical query in the search, improving the relevance of the searched documents and increasing the probability of achieving what the user wishes to get.

Keywords:

Concept, HDWiki, HowNet, Information retrieval, Logical symbol, Query formulation, Semantic similarity, Ternary relationship, Wikipedia.