Research On WEB Semantic Structure Based On Self-consciousness

In order to make people find the information more quickly and accurately online,based on the semantic web,combined with the theory of dynamic ontology and self-consciousness of agent,the paper proposed the concept of WOW agent and the idea of the semantic web based on self-consciousness.The overall framework of this semantic web,which describes its work flow,and the composition structure of its basic elements-the website agent,the user agent and the WOW agent.The WOW agent is the core of the semantic web based on self-consciousness.It has the preference information of the user when them online,and has a comprehensive understanding of the dynamic ontology of the Internet composed of many website agents.Therefore,the WOW agent can provide users with personalized service according to user's preferences and the current state of the Internet,so that users more convenient and effective online.The web semantic structure refers to the The WOW agent's understanding of the current state of the Internet,which mainly includes the WOW agent's understanding of the website agent,the the information object and the classification of the relationship between the website composition.In order to explore the classification relationship of websites in the Internet and design the web semantic structure,the paper proposed a clustering algorithm of sorting on website's characteristic value,and thought this algorithm we divided more than 30,000 websites which randomly acquired into 300 categories.the algorithm can be divided into three stages,in the stage of subject words collection,through the crawler we crawl million web pages,grab the subject words on them and streamline optimization;in the stage of the website eigenvalue acquisition,he structure of the eigenvalues of the website is designed,the keywords are assignedand assigned,and the eigenvalues of each website are stored in the form of files;in the stage of website classification,use the idea of radix sort to sort the website,and then calculate the distance between adjacent websites,according to the size of the distance value of the website classification.Finally,according to the results of website clustering,the WEB semantic structure is designed.