Data Mining For Applied Research In Information Management Systems

As the information of website is more and more abundant as well as the topology more and more complex, there exists the situation of "information overloading" and "resource maze" in information service. Today the face of a broad array of information, people are concerned about is how to effectively collect and select information of interest, and, more importantly, people are concerned about how the growing number of information automatically discover new concepts, and automatic analysis of the relationship between them , can truly succeed in the automation of information processing.Accurate and timely grasp of users on the system's visit is to ensure that efficient system to achieve an important basis for adaptability. This paper mainly on the status of data mining technology and development, improving the k-means algorithm, k-means algorithm optimization of the initial category centers, each of the Centre and to improve the method of calculating the last time this category centers and the iterative the average of all cluster targets as a means of seeking re-centre, reducing the impact of noise, speed up the convergence of the iterative function to enhance the efficiency of the algorithm, and applied information management system. According to the system log data users access to the various modules of the degree of support to closed or open information module webpart to achieve site intelligent, automatic system module to provide recommended changes users view and optimized system structure, improve the quality of service information site, promote the development of Intelligent Information Processing field. It is important research significance in theory and in practice.