|
In recent years, data mining (Data Mining, DM), extensive attention by international artificial intelligence and database community. However, with the advent of the Internet age, the traditional data mining objects change, new challenges for data mining and knowledge discovery, Web mining precisely this context. Web mining is identified from a variety of data in the Web world valid, novel, potentially useful, and ultimately understandable patterns process. Web mining has become an important Web information decision means Web usage mining because of its convenience and accuracy of data mining is to become one of the important research direction in Web mining. China's Internet has been very popular, one of the primary means for people to obtain various information. The Internet and the real economy continued integration, using the Internet to transform and upgrade traditional industries and the traditional industrial restructuring and the transformation of the mode of economic growth, the Internet has become a new strategic industry for the development of low-carbon economy in China. The Internet industry data released by the Ministry of Industry, as of the end of 2009, the number of domestic websites reached 3.23 million, an annual growth rate of 12.3%, the number of Internet users reached 404 million, the information industry accounted for the proportion of the gross domestic product (GDP) reached about 10%. With the continuous development of the Internet industry, the competition between the site of the white-hot level, how to stand out in the increasingly fierce competition in the website is the main problem facing policy makers in. \This requires website operators to understand the feel of the user for the site visit, at the same time according to the user's needs for the site in a timely manner reasonable improvement, and thus to win the favor of the user. The log file is the site direct access to the most comprehensive record of user access, user access to process all the information recorded in the log file. Web usage mining is the Web log files are found habits and access patterns of user access, so as to optimize the the running layout and structure of the website, to enhance the site user satisfaction. This paper Jiangsu Entrance Examination network real operating data, Web usage mining technology to conduct a comprehensive analysis of the mining log files for the operation of the site, to discover the habits and access patterns of user access, and then found the site running status page the relevance, timing, and final optimization strategy to help decision makers in the mining results, has a valuable practical significance for the website to adapt to future trends, and to accelerate its own development, and to respond to competition and challenges. Papers innovations mainly: a comprehensive sort of theoretical knowledge of Web usage mining; entire process for Web usage mining depth, specifically the main problem in data preprocessing solution; theoretical research based on the integrated use of computer technology, database technology, data mining and other means, to establish a \
|