Libraries at University of Nebraska-Lincoln


Date of this Version


Document Type



This research is to analyze the data mining by analyzing clusters formed. This study was about determining the number of libraries available at the primary and junior high school levels in Indonesia. The main role of libraries is in the current implementation of quality education. The data are from the central statistics agency which contains 34 records indicating the number of libraries at the primary and junior high school levels. The proposed method is a combination of the C4.5 and k-means methods. clustering is performed based on the number of clusters (k). Thus, the results will be classified according to the C4.5 method. Using the number k = 2, the clusters obtained have average centroid values. (10786.75 and 3210.25). Meanwhile, there are thirty provinces in the low cluster (K-2) with centroid values . (1744.2 and 626.9). The results of the cluster group formed using the standard parameters (criterion = gain ratio; maximal depth = 20; confidence = 0.25 and minimum gain = 0.1) was accurately classified 97.5%. A combination of the C4.5 and k-means methods should be used