Research On Construction And Application Of Ontology Based Knowledge Base In Inner Mongolia Autonomous Region's Tourism Domain

With the development of Internet technology,data has grown exponentially.Meanwhile,search engines have become an important tool for people to acquire knowledge quickly.How to standardize the massive Internet data and exploit useful information from Internet is an important subject in the field of artificial intelligence.Since Google proposed the concept of Knowledge Graph in 2012 and successfully applied it in search engines.Knowledge Graph(also called Knowledge Base)has become a technology to establish and standardize a certain scale of knowledge.After that,the academia is focus on how to build vertical knowledge graph in specific fields.Among them,more knowledge graph in the financial domain,fewer in the tourism domain.And tourism domain knowledge graph in Inner Mongolia's have not yet been published.This paper integrates the tourist attractions information in Inner Mongolia Autonomous Region and establishes a domain ontology knowledge base on multiple data sources.Firstly,we design a crawler tool to obtain information about the A-level attractions in Inner Mongolia Autonomous Region from the encyclopedia Websiteand the tourism Website.Secondly,generate the three-triple knowledge <entity,attribute,attribute value> and <entity 1,relationship,entity 2> through the pre-processing work such as knowledge cleaning.Third,edit the acquired data through the Protégé that ontology editing tool to standardize the concept of the pattern layer from top to bottom,reducing the cost of docking with the data layer.Finally,construct the knowledge base through saving the ontology in pattern layer and the three-triple knowledge in data layer by using the graph database Neo4 j.The returned results for the user's question of the existing search engine are mostly document-based sorting.These ignore the semantic relationship of natural language with poor accuracy and no intelligence.In order to apply the constructed travel knowledge base,this paper realizes the QA function and designs a Web interface.Knowledge base question answering use the methods based on matching template,the main process is when user asks a question in natural language,the system will goes through the three steps as question classification,question analysis and return answers,then returns the result to user finally.