Establishment And Search Of National Security Threat Knowledge Base Based On Social Network

As a new kind of media of Web2.0,there are a huge number of users in the process of information manufacture,dissemination and consumption of microblog.With the increasing popularity of online social platforms such as Weibo,online social networks have accumulated a large amount of data generated by users.How to acquire information from these huge amount of data to meet the needs of users is a problem to be solved urgently.Although search engines have undergone many improvements over the past few years,the query processing techniques still make use of the keyword-based symbol matching method,and the query results can hardly meet the needs of users.Query expansion is a better solution to this problem.The main work of this thesis is as follows:(1)Propose the method of terminology identification and extraction in the social network security domain,identify and extract the terms of microblog data.Using conditional random field model,combined with active learning strategy,more than 500 terms in security domain are extracted.As the number of training samples for active learning increases gradually,the performance of the model also increases.With the introduction of active learning strategy,the accuracy rate of the corresponding training set and the increasing rate of the accuracy with the growth of the size of the training set increase significantly,compared with the non-active learning strategy.The term recognition efficiency has reached a higher level.(2)Propose the method to establish and expand the national security threat knowledge base,design the ontology establishment process of security domain and use Protege 5.1 development tools to establish the security domain ontology.With the ontology expansion method based on inter-semantic relevance,the ontology of security domain is expanded according to the semantic relevance rules.The expanded ontology contains two categories,six subcategories,25 classes and some examples of security events,including natural disasters,public safety events.(3)Propose a microblog query expansion algorithm based on ontology and local query feedback(OFQE),and a microblog query expansion algorithm based on ontology expansion and Borda count rank(OBQE)for big data search of national security threat in social networks.Take the short texts of microblog as the research object,combine with the ontology semantic expansion and query feedback,and semantically expands the initial query by establishing the security domain ontology.Modify the query expansion words according to the local query results for secondary query search,and finally optimize the rearrangement of the query results by the Borda count rank to improve the microblog search performance.(4)Implement the social network big data search system based on national security threat knowledge base.The functions of the system include information extraction,ontology expansion and data search.And the effectiveness of the system is verified through testing.This thesis establishes the national security threat knowledge base,and do some research on the social network search based on knowledge base.Through the establishment of the national security threat knowledge base,we can provide accurate search services for online social network search combining with microblog query expansion technology and Borda count rank,help people get more valuable information from social networks and realize the search services to meet the individual needs of users.