Clustering is a process to group data into several clusters or groups so the data in one cluster has a maximum level of similarity and data between clusters has a minimum similarity. X-means clustering is used to solving one of the main weaknesses of K-means clustering need for prior knowledge about the number of clusters (K). In this method, the actual value of K is estimated in a way that is not monitored and only based on the data set itself. The results of the study using the X-Means algorithm with the Davies-Bouldin Index evaluation to determine the number of Centroid clusters is done by modifying the X-Means method to do some centroid determination to get 11 iterations. The result is produces cluster members that have a good level of similarity with other data. In determining the number of centroids, use the Davies-Bouldin Index method where testing with 2 clusters has a minimum value with a DBI value close to 0.
The clusters number optimization problem is a problem that still requires continuous research so that the information produced can be a consideration. Cluster evaluation techniques with Sum of Square Error (SSE) and Davies Bouldin Index (DBI) are techniques that can evaluate the number of clusters from a data test. Research with these two techniques utilizes Stunting data from a number of regions in Indonesia. The result is information on stunting data which is formed from the optimal number of clusters where the largest SSE is formed at k = 5 and the smallest DBI is formed at k = 5, with values of 23.403 and 1,178 respectively. Changes in the number of clusters also influence the information produced and DBI is proven to produce optimal number of clusters that contain information with a better pattern because it has a small intra-cluster value. It is expected that the results of this study can show the performance of the two evaluation techniques in producing the optimal number of clusters so that grouping information is in accordance with the expected pattern.
The study aims to analyze the selection of exemplary teachers using C4.5 algorithm, which is one of the existing decision tree methods in data mining theory. Teacher data was obtained from the school in SMA Negeri 2 Pematangsiantar (2010-2016). The data used contains information about teacher history and teacher assessment data. This research uses interview technique and questionnaire in obtaining data. There are 11 attributes used in the assessment process: NUPTK, name, age, education, status, Appointment Letter, competence, award, work, teaching load, personality, Position, and label. The system has been tested using the Rapidminer application with 16 data samples. Where the rules are obtained as many as 24 rules. The level of accuracy of the system states that four of the ten attributes have a very significant correlation to the model of relationship rules in determining the proposed model teacher, such as (Position, competence, education, and personality). The four attributes (Position, competence, education, and personality) contribute 82.8% to the model of the rule of connectedness rules in determining the best teacher.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.