Data Mining Applied Research In The Library User Behavior Analysis

Many service conceptions are changing with information times coming.As a industry special nonprofit public service new requirements are raised to service mode and conception of library.One of the feasible and effective methods is to construct the analytic system with data warehouse and data mining.The function of data mining is to extract implicit,previously unknown,and potentially useful information from database or data warehouse.The information will supply strength instruction for people to analyze and make decisions.The system could facilitate the management of library.The paper tells us how to combine data mining technology with the management of reader's data and to extract useful information from mass data.It tells the knowledge of data mining and careful analysis of the relative contents. and decides to apply the method of decision tree and sequence analyzes mining to the system of reader analysis.Firstly,decision tree is applied to analyze the number of reader to find out the factors that may affect the number and the grade of affect.Make use of the number to measure the reader's loyalty,and to predict the types of readers with the consequence of decision tree analysis,so that to supply decision-making information for administration department.So as to promote to supply different service to different types of readers and make good advantage of library resources. Next,sequence analysis mining is used to analyze the reading habit and to find the rules of reading.So that librarians may provide personalized recommendation or recommend readers with similar needs the reading information.At last,a summarization is made and other suggestions for improvement are raised according to the shortcomings now exist in the system.