Research On The Application Of Data Mining Technology In The Utilization Of University Library Resources

With the continuous progress and development of data mining,the application of data mining technology has involved all aspects of people.Data mining of university library borrowing data can grasp readers 'reading trends in general and make scientific predictions based on the trends,thereby improving the management level of college libraries and readers' reading atmosphere,which is beneficial to the construction of library collections.This paper mainly conducts mining analysis from two aspects.First,cluster analysis of borrowed data and analysis of reader and book clustering characteristics,so as to provide more rigorous and scientific research methods and ideas for library service innovation and library cooperation.Second,analyze association rules on borrowing data,analyze association borrowing information using association rule mining technology,accurately determine the borrowing behavior of readers,and understand and grasp the characteristics of interest,potential needs and actual needs of different professional reader groups at different grade levels.The data mining tool selected IBM SPSS Modeler,the clustering algorithm used the K-Means algorithm,and the association rule algorithm used the Apriori algorithm.Because these two algorithms are well embodied in the above tools,they are very suitable for studying library borrowing data,and the accuracy and accuracy of the final mining results are also guaranteed.Finally,this paper proposes an improved collaborative filtering algorithm to implement the book personalized recommendation function,and prototypes the front-end and back-end of the book personalized recommendation module.The front-end is mainly a personalized book recommendation interface,and the back-end is mainly a system management interface and a popular book management interface.