Automatic Discovery Technology Based On The Hot Topic Of Multiple Data Sources

With the rapid development of Internet and network technology, the network has become anemerging media and channels for people to get information. Faced with exponential growth ofinformation and data on the Internet, how to obtain the information from the vast ocean ofinformation, which is needed and get interested, is also an issue of common concern. Hot topicdetection technology is the practical application of the topic detection and tracking technology.It is possible to find a hot topic from the information flow on the network, which can helppeople to understand and realize the events more comprehensively. The technology has greatpractical applications in government, finance information security and many other areas.This article firstly summarizes the research status of the topic detection and trackingdomestic and abroad, and then introduces hot topics detection as well. Analyze and summarizethe problems faced and needed to be improved. To solve these problems, this article makesprincipal research and improvements. The main work done in this article is as followed:Firstly, the paper considers both from the media and users, mixes the integration of bothcharacteristics and proposes a heat calculation formula to make assessment of the topicsaccording to the news reports and microblogs. By utilizing the heat calculation formula to getheat value to assess the hot topics, and then sort the hot topics. Finally, it can get a hot topic atany time according to the heat value of the sorting. It is convenient for people to keep abreast ofthe latest, hottest topics, to help the government departments to monitor and to guide onlinepublic opinion.Secondly, the paper makes some improvements on topic detection algorithm, and proposesa new algorithm of hot topic detection based on keywords. Define the keywords, and use thekeywords to represent the topics. In the improved algorithm, it adopts two clustering strategy.Make the first clustering of the news`title to find the newly emerging topics and set the initialthreshold, combing the eligible reports into its corresponding topic sets. Then make the secondclustering of reports of the topics, use heat calculation formula to calculate the heat of hot topicsto make further analysis. Finally, make experiments with specific case to verify the feasibilityand the actual application of the algorithm and thought put forward in the article.Thirdly, the heat calculation of hot topics and the approved hot topic detection algorithm areapplied to the network public opinion analysis and monitoring system. Make the overallintroduction of the system, describe the system function and the realization of each module, andfinally make the experiments by combining the actual case.