Implementation of Data Mining Clustering Using the K-Means Method in Grouping Library Books

Abstract

The K-Means algorithm is an iterative grouping algorithm in partitioning the data set into a number

of clusters that have been set at the beginning. The formulation of the problem in this study is how to apply Data

Mining with the K-Means clustering algorithm to find solutions in adding types of reading books that students

are interested in. The application of the K-Means Algorithm aims to assist the librarian in classifying the data on

borrowing books that students really like or are interested in. In the k-means algorithm, it can analyze more

deeply by processing book lending transaction data with results that explain various information - information

about the distribution of the intensity of borrowing a book so that the information from the processing results

can help the library to determine the addition of a collection of books that are right on target. which is then

processed using Rapid Miner. This provides benefits for the Politeknik Negeri Balikpapan Library in the plan to

increase the collection of reading books in the library. The final results of the research are 107 book titles with a

borrowing frequency of 231 times in the period 2019-June 2022. The test is carried out by finding the smallest

value of the Davies Bouldin Index (DBI) where after the data is processed it is known that the smallest value is

0.407 with a total of 3 clusters. It was concluded that scores with the Highest, Medium and Low Clusters were

obtained in the grouping of library books, namely the Highest Cluster with the number of book data being 2

titles, namely Teknologi Beton, teori dan Praktik Hotel Front Office. These two types of books are the most

borrowed so that they can be recommended to be proposed in the procurement of library book collections.

Medium cluster with 23 book titles and low cluster with 82 book titles.