Analysis of Self-Organizing Maps and Fuzzy C-Means methods in Clustering Teacher Data for Nominations of Candidates for Education Unit Supervisors

The quality of education is very dependent on the management of education management one of the important factors in the management of education is monitoring and evaluation. Therefore, continuous Education Supervision supported by the appointment of competent supervisors will have implications for the quality of education. Supervision of education management is carried out by the principal or teacher appointed to carry out managerial and academic supervision in certain educational units. The problem that often occurs is that grouping teacher data to be selected as supervisor candidates are still conventional. Therefore, a teacher data grouping model is needed to obtain useful information in planning strategic steps and policy regulations for determining prospective supervisors for academic units. By utilizing Information and Communication Technology, especially in the field of Data Mining, this Teacher Data Grouping research uses the Fuzzy C-Means and Self Organizing Maps method, and the grouping results are analyzed by measuring the spread of data on each grouping formed by using cluster variance. The output of the Teacher Data grouping process using the Fuzzy C-Means and Self Organizing Maps methods can bring up a group of Teacher Data nominations that are competent to be selected as candidates for supervisors for certain educational units. The research results were obtained by forming several groupings in Fuzzy C-Means by providing an error accuracy value of 0.1 and Self Organizing Maps, which is set by the learning rate and learning rate; the results obtained are grouping with 3 clusters by providing a learning rate of 0.8 and a learning rate of 0.7 in the Self Organizing Maps method has a Variant value that is ideal compared to grouping on Fuzzy C-Means and rather than the same method by forming different groups.