Data mining in academic libraries using the K-means algorithm

Abstract

Data mining is an complex process that extracts the required, useful, and comprehensive data from the
large quantity of data in order to achieve predetermined goals; some consider data mining to be a
common term in the field of extracting knowledge, while others consider data mining to be one of the
first steps in the process of extracting knowledge. And the clustering is the method of separating a data
set into distinct cluster groups, each with comparable characteristics, with the goal of combining the data
into a single cluster. We are attempting to discuss the possibilities provided by the data mining procedure,
and well as how it might improve the standard of service in universities libraries, in this study. Where this
article goals to set the books information that is contained in any library, for example, a library College of
Education for Pure Science in the University of Thi Qar, by using the K-means clustering approach.
Name, Title Of the book, and Author are used as input variables in this K-means clustering method. The
result is divided into three groups: (B1) the most commonly borrowed book, (B2) often borrowed book,
and (B3) rarely borrowed book. The final result achieved using this K-means Clustering approach
includes members of cluster 1 containing (19) members, cluster 2 containing (22) members, and cluster 3

containing (19) members. The library can use the information from grouping this books data. In the case
of selecting books to be introduced to the library, it is important to reduce the number of books that are
seldom borrowed for the purpose of avoid a backlog of books that are seldom borrowed, leaving room for
new books to also be added.