Research And Implementation Of Named Entity Retrieval Based On Ontology

With the rapid development of the Web technologies, search engines hava become the main way to access web information. However, the traditional search engines return more monotonous results. Usually conventional search engines return a list of Web pages ranked according to their estimated likelihood of relevance to the query, but it often ignores named entity information. Therefore, named entity retrieval has recently become an important search task in Information Retrieval. The goal is not to find documents matching query terms, but, instead, finding named entities. Currently, there are not enough abilities to understand in queried semantics in the retrieval of named entities. In this paper, ontology, other information retrieval techniques and the named entity retrieval are combined together, implementing the understanding of the query semantic.In this paper, we firstly analyze current named entity retrieval technology and its theory, and we propose a formal model to search named entities. Secondly, in this paper the English Wikipedia has been analyzed in accordance to its articles and the category link graphs to extract a English pseudo-ontology, and we use this technology to extend ontology antomatically. We extend the classical Vector Space Model representing named entities in it as a weighted profile, and find the coordinates of a named entity in the vector space, resulting in named entity vectors in the same vector space as documents. Lastly, different types of refinements can be applied in order to include more evidences of relevance, and we integrate ontology-based expansion algorithm into the model.We experimentally evaluate our system on the expert search task in order to show how it can be adapted to different scenarios. Ontology-based expansion method effectively solved the ambiguity of user queries, a combination of several extended algorithms makes the system’s MAP improved significantly.