The Design And Implemetntation Of Hot Events Mining System

With the development of Internet and information industry,especially the rapid development of intelligent mobile terminal,more and more people can hardly do without the Internet.The real-time,fast,open,the network has changed from a large extent and affect people's habits and ways of life.People get news from the Internet,access to information and entertainment,access to professional consulting and so on.Due to the progress of technology,the rapid development of science and technology,the rapid expansion of information,the price of intelligent terminal is lower and lower,as well as the intelligent terminal equipment's function has improved very fast.Get information by intelligent terminal has become a mainstream way.When a hot event happened,people can browse and search the event information through the handheld intelligent terminal e.g.phone,pad and so on.Many of the current hot events study is based on Weibo,RenRen,BBS,Twitter and other social networks.And here we are ready to use the log of search engine which was logged through the browse of their intelligent mobile terminal to mine the hot event and find its influence to total network traffic.We hope the platform can learn the user of search engine at the level of event.The platform will get sudden query,and then clusters the sudden query.The platform will mine the hot event through sudden query cluster result.First,the thesis introduces the related theories and technical background of the mining of hot events.Second do requirement analysis and system design for the hot event mining system.Third we will get sudden queries,then we will introduce how to calculate the similarity of queries.Then we will cluster sudden queries the through the method.After that we will use the cluster result to recall high frequency queries,and then we will do something about data visualization.At last the thesis will summarizes the whole work of the system and looks forward to the future work.