Research Of Semantic Ontology-based Vertical Search Engine Model

With the popularization of Internet and the increase of information on the Web, more and more users tend to use search engines to obtain information. Currently, there are two main types of search engine:general search engine and vertical search engine. Vertical search engine is a new service pattern against the general search engine, because general search engine contains larger amount of information and the accuracy is not high. The vertical search engine is the inevitable trend of search engine industry.However, the vertical search engine still uses keywords based retrieval method, which can not satisfy the needs of semantic search. The main reason that exerts this problem is that documents are lack of semantic annotation. Therefore, search engine could not do semantic analysis on user's query. But ontology can accomplish information annotation, realize semantic search. Combine ontology with search engine becomes an important research means of semantic search engine.Therefore, this dissertation focus on the construction of the domain ontology, ontology-based information extraction of structured data and semantic query expand method using the domain ontology. Then, introduce a framework of ontology-based vertical search engine, design and implement a prototype. The main works of this dissertation are as follows:(1)Build a catering ontology EnCatering by analyzing the catering information, which includes catering class, dishes class, cateReview class and location class, then define the attributes and relations between classes in order to better realize knowledge representation and information organization.(2)According to the features of catering websites, design rules to extract the ontology individuals automatically, which make foundation for query expand and analysis.(3)Create index for classes, attributes, relations and individuals, use the index to implement semantic analysis and query expand, emphasize keywords-ontology entity matching algorithm and the process of semantic query expand.Finally, develop a vertical search engine Catering Search based on constructed ontology. Experiment shows progress on overcoming the problem in keywords-based retrieval, suggests that ontology-based vertical search engine is practical significance.