Design And Implementation Of Financial News Hot Information Mining System

With the development of Web 2.0, the amount of news on Internet is increasing at a sharp rate. For everyone, news is the source of acquiring information in daily life. The rapid growth of information gives us in-depth understand of all aspects of life. News is very helpful in specific industries because it provides important reference. As for financial industry,understanding of current events may become important reference for investment. The investment and forecast in economy field are based on development status and future development trends of related industry.Understanding of the trends of industry is helpful to provide direction for economy decisions. However, under the situation of rapid increasing information, traditional online news displays start to show their disadvantages.Traditional gate sites display news by dividing them into separate sections,where each section contains a list of title of related news. While browsing the site like that, the user may get board in the reading progress and gradually lose interest to continue reading. In order to let user understand the latest news quickly and accurately, it is valuable to study the extraction of hot information from news text.In order to solve these problems, this paper proposed a financial news hot information mining system. It takes the advantages of data mining into financial news analysis to achieve key terms extraction and text clustering.Based on data obtained from extraction and clustering, daily news hot words are extracted in different time period and hotness value of each news is calculated. Thus the hot information of financial news is obtained. In order to provide convenient visual representation, the data representation in this paper is designed and implemented based on Web technology. The page presents news hot information using tendency chart and word cloud. In order to increase the efficiency of browsing and take the advantages of different type charts, the charts on page are interactive.This paper consists of three parts. The first part is about the requirement analysis of financial news hot information, news storage and data process. It includes the design of storage schema of financial news, and the data pre-processing based on whole application requirements. The second part aims at financial news hot information extraction. It includes the application of data mining methods based on intermediate data obtained from part one.The results of part two are part of the data for display. The last part is design and implementation of the final representation using data obtained in previous two parts.