Research On Improvement And Application Of Tongyici Cilin

In today's era of machine intelligence,the research and application of artificial intelligence has become a hot spot.From artificial intelligence theory itself,or the technology and application of artificial intelligence,all need to solve a problem,is to use knowledge to describe the ability of intelligent behavior.In order to solve this problem,the first thing you need to the representation of knowledge,to join the composition of semantic knowledge,then you need to set up corresponding knowledge base.In the traditional database,with the deepening of the database theory research,in order to overcome the shortage of database model in expressing ability,strengthen the composition of semantic knowledge,has the reasoning ability,and then puts forward the concept models of a higher level of abstraction,use the corresponding knowledge expression,that is a great similarities with the artificial intelligence in the knowledge base system.Moreover in natural language processing,the level of the natural language understanding,including lexical analysis,syntactic analysis,semantic analysis and pragmatic analysis.The ultimate goal of natural language understanding is a perfect pragmatic analysis,and semantic analysis is the basis of pragmatic analysis.To analyze the semantic to knowledge representation in the first place.Now the knowledge base system has now been in expert system,decision support system,office automation system and so on many fields has been widely used.Knowledge base mainly can be divided into two kinds,one is based on the field of domain knowledge base,is a kind of common sense knowledge.These two kinds of knowledge base is the two basic knowledge of computer systems.Early in the artificial intelligence,expert focused mainly on domain knowledge base,and now experts also began to focus on common sense knowledge base,is a kind of artificial intelligence and knowledge engineering technology progress.In the study of artificial intelligence,common sense knowledge is collected facts and a human being to predict information.Aimed at creating a commonsense knowledge base:A database,including knowledge,these knowledge are most people know.Expressed by the way,make its can be used in artificial intelligence applications of natural language or make ordinary the inference of the world.This paper is mainly on the basis of the existing common sense knowledge base,mainly for the extended edition Lin synonyms words do progress improvements,and to do the word relevance computation.This paper based on the commonsense knowledge base WordNet,HowNet and Tongyici Cilin for forest research,on the basis of Tongyici Cilin,combining the WordNet and HowNet.In this paper,it is to improve the Cilin.Cilin when doing the word similarity computing is more reasonable.Based on this,this paper made the following content:(1)Introduction of HowNet,WordNet and Cilin.The classification of words,mainly introduces the internal structure of the knowledge base.Mainly introduces the original meaning and meaning in HowNet,the relationship between the concepts.Mainly introduces the establishment and expansion of Cilin.(2)Cilin improvements.According to the description of the relationship between words and words in HowNet and WordNet,add these relationships based on Cilin.There are hyponymy,Hypernyms and Hyponyms,Antonyms and so on.(3)Based on the improved word Lin do word relevance computation.When calculating the semantic similarity between words,it is taken words correlation into account,thus increasing the rationality of word semantic similarity value.