Grouping community reading interests using the k-means clustering method (case study: Magelang district library and archive service)

Abstract

So far, the problem of gathering information at the Library and Archives Office of Magelang Regency is relatively low. To increase reading interest, policies are needed to determine reading interest. So the data used is book transaction data. This study aimed to classify people's reading interests according to the number of borrowed books at the Magelang Regency Library and Archives Service using the K-Means Clustering method and to find out which book categories are most in demand by the public at the Magelang Regency Library and Archives Service. One way to manage this data is to use data mining using the K-Means method. The results of this study are low reading interest, evidenced by using 2 clusters and the category of books that are most in-demand Literature with a high cluster strength value, namely with a Silhouette Coefficient value of 0.7354.