The Research Of Reading Tendency Of College Students Readers Based On Data Mining

Since the 21st century,the development of the society accelerated,readers are affected by all aspects of society so that reading tendency of them is changing,especially for the college student readers who equipped with strong ability to accept fresh things,the changing cycle of reading tendency shortening.As the service institutions of higher education,university library should fully understand readers' reading tendency and grasp the readers' reading interests,which directly affects the service quality of university library.Nowadays the library management system lacks the function which can analyze readers' reading tendency with the borrowing and returning data.Attaining the valuable information from large volumes of data,data mining is a better way to solve this problem.This paper summarizes the research status of data mining and reading tendency,the overseas and domestic research status compared.Meanwhile,this paper discusses the definition of data mining,tasks,processes and methods of data mining.The methods of data mining mainly include clustering analysis,association rules and predictive modeling.Based on this basis,this paper serve borrowing and returning data of one class graduates of one university as the research object,completing the work of data preprocessing by SQL Server,such as data cleaning,data conversion and data integration.In phase of data mining,this paper mainly analyzes the readers' reading tendency from two aspects:On one hand,this paper studies the readers' reading tendency by cluster analysis.First of all,it clusters readers based on the number of readers' borrowing and the interest by K-Means clustering algorithm,and analyze the characteristics of every kind of readers,which lay a foundation for association rules analysis.Secondly,it clusters books based on the number of borrowing and analyze the composition of each kind books,which can provide library with suggestions in respect of book weeding and books purchasing.On the other hand,this paper studies the readers' reading tendency by association rules analysis.First of all,it analyzes the relation between different CLC by association rules analysis based on the result of clustering analyzing,which makes us grasp the readers' reading tendency of different stages in university.Readers have a chance to enjoy better recommendation service.Secondly,it analyzes the relation between schools and CLC and the relation between majors and CLC by association rules analysis,which can grasp the readers' tendency in different schools and different majors.Based on the result of data mining,this paper provides some suggestions for the library.Finally,all the work and research results are summarized.The shortage and further research work are pointed out.