APPLICATION OF IMPROVED K-MEANS CLUSTERING ALGORITHM AND SIMPLE ADDITIVE WEIGHTING (SAW) METHOD FOR DECISION SUPPORT SYSTEM OF TEACHER PERFORMANCE ASSESSMENT AT SCHOOL A BANJARMASIN

• Main Author: Mawarni
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/68976
• Institution: Institut Teknologi Bandung

School A Banjarmasin is a boarding school where the principal motivates its teachers by selecting the exemplary teacher and increasing the salary which are divided into 4 clusters based on teachers' performance. According to the results of interviews, in 2019 and 2020, the principal did not assess teacher performance appraisal, so the raise in the salary was equalized. There was no backup data of teacher performance from the previous principal, and there was no system available to assess with the performance appraisal. The author proposes to design a Decision Support System (DSS) that aims to support the decision-making process using the Improved K-means Clustering algorithm for clustering salary increases and the simple additive weighting (SAW) method for determining exemplary teacher. The programming language chosen for designing the DSS was Visual Basic for Application (VBA) Excel. The data used in this study were the results of teacher performance assessments in 2021/2022 with 4 assessment dimensions consisting of 11 assessment criteria. There are 2 outputs in the DSS, output 1 and output 2. The results shown in output 1 has 97.30% accuracy based on the provisions of the school. There are only 3 clusters that have members while cluster 4 has no members because there are no data that meet the cluster requirements. The results shown in output 2 are the clustering recommended by the author where there are always 4 clusters of salary increases. The DSS designed will be very helpful for decision makers because it delivers calculations and rankings that are more accurate and objective. It also provides recommendations to be considered by decision makers in the future.