| In this information society, as a kind of resource data already received more and more attention. With the rapid development of Internet technology and the wide application of the database, the human society has produced vast amounts of data resources, and the data continues to expand. How to dig out useful knowledge from vast amounts of data is a problem placed in front of scholars, so the data mining come into being. This paper first details the basic concepts of data mining. Then, for the issues to deal with, the paper in accordance with the entire process of data mining, by taking a certain approach, mines Internet enterprise data. And the paper analyzes the mining results. Finally, the paper developed a data mining system based on GIS.The paper focuses on obtaining Internet enterprise data and data mining. The method of obtaining data includes automatic acquisition and non-automatic acquisition. Some attribute data of Internet companies are obtained by automated means, that is to say crawler. Data in accordance with the issues to be studied are divided into two categories. For the study of Internet companies macroscopic distribution, the paper selects the Internet companies whose main business is website construction. In addition, for the study of urban network structure and the property features, the paper selects Internet companies which are listed on Shenzhen Stock Exchange. Among them, the study of the properties characteristic is mainly about Internet companies in various fields, the method used being the classical association rule model as well as traditional statistical methods. But mining spatial data is based the cities of enterprise distribution, the model used being the "chain of World city network" (IWCN) model. The article is divided into seven sections. Finally, the paper uses C# and ARCGIS Engine10.0 to develop a mining system on the Internet enterprise data. The system can get Internet companies data online, and achieve data review and graphs show and other functions.Through data mining of Internet companies, the paper has the following conclusions:First, the spatial distribution of Internet companies shows core-edge distribution. There are four main gathering area, namely:the Pearl River Delta, Yangtze River Delta, Beijing, Tianjin Strip and Xiamen, Fujian area. Second, on the study of the characteristics of the Internet business property, companies time to market 7--10 years, leaders between the ages of 48-55.5 years old, master’s degree are more closely linked. The time to market from 0 years to 19 years is before decreasing increments. The most of leaders are masters. Leaders aged are between 40-49 years and 50-59, and reached a considerable proportion. The male leaders occupy the main location. Third, in the research network structure of the city, for the Internet technology and Internet enterprises+ enterprises, urban connectivity are the Beijing highest. In the city connectivity, Web presents "diamond" status. The biggest difference is the Internet+enterprise "diamond" structure is more full, presumably Midwest area focusing on use of Internet technology to improve the traditional business model. Fourth, the article develops a data mining system for the Internet companies based on GIS. The system is mainly used to obtain the relevant data of Internet companies through crawler, such as the chairman of the board of the age, education, gender, place of origin, etc. And the system can view the data by maps, various types of charts, mapping. |
PDF Full Text Request