The Design And Implementation Of Ontology-based Medical Search Engine

With the popularization of Internet and the increase of the medical information on the Web, more and more users and medical personnels incline to use Internet to collect medical information. Currently, users mainly use general or medical search engine to get medical information. General search engine does not process the medical information properly and cannot be qualified for medical information search. However, currently most medical search engines based on key words are not efficient enough to satisfy the needs of the users: firstly, they can not make use of the documents semantically; secondly, they can not provide semantic analysis on the user's search query and only match the query words with the documents. Medical domain ontology describes the concepts and relations of this field on the semantic level, which includes well defined concept hierarchy and logical reasoning ability. It is expected to provide semantic support for user's semantic search.This dissertation, based on the research of the construction of the domain ontology, introduce a multi-information integrated semantic search method by automatically extracting the ontology instances and the semantic processing of user's query. A prototype system has been designed and implemented.Based on the analysis of the related technologies and the requirement of the ontology-based medical search engine, the main work of this dissertation is as follows.1. The International Classification of Diseases has been converted into the disease class in a medical ontology; also the hospital class, drug class and doctor class have been constructed by defining the concepts, attributes and the relations between them.2. Semantic analysis on the user's query provides the base for semantic-based information retrieval. The semantic analysis and query expansion have been implemented by establishing the index for the concepts, attributes, instances and relations of the ontology.3. According to the features of medical Web sites, information extraction technology based on the rules has been designed to automatically extract the ontology instances.Based on the research above, a prototype system of ontology-based medical search engine has been designed and implemented, which could provide the related information to the user and shows the progress on the overcoming the problems faced by the medical search engine based on key words. These works demonstrate the importance of ontology on the semantic search of multi-information.