Website evolution based on statistic data

Maintenance and evolution are critical for website since the requirements often change, the developing cycle is short, while the life cycle is long. In order to make progress in this area, we focus on the users´ responses and attitudes. So we begin our work at the log files on the side of website server, gathering the users´ visiting information and the server´s responses. Thus we obtain our weighted structure model. Then, we analyze and discuss such information to identify the key pages, the predominate pages and the users´ visiting patterns. And we can improve the testing efficiency based on these knowledge. All these will help to improve the structure of the site, fulfill the functionality of the site, and enhance users´ visiting efficiency.