Analysis Of Employee Discipline Based On Digital Attendance With The K-Means Algorithm Method

Abstract

Employee discipline is one of the most important factors for the progress of the company. PT. Sumatra Core Cellular (PT. SIS) Pekanbaru has implemented a digital attendance application, but the company has not evaluated the application to determine the level of employee discipline. Data mining is the process of extracting useful information from a large database population. One of the data mining methods is the K-Means algorithm. The data mining process uses the method of K-Means algorithm with 2 clusters namely discipline and less disciplined categories. The data used is attendance data of 159 employees, namely data on tardiness, non-attendance (TAP), attendance hours and 4 selected questionnaire questions. Tools for grouping with the Rapidminer application. Using the K-Means algorithm method, it is known that cluster 0 consists of 133 employees or 83.64% with a disciplined category and cluster 1 produces 26 employees or 16.35% with a less disciplined category. Judging from the accuracy of attendance hours, employees in cluster 0 are more likely to be present at 07.45 - 08.15 and in cluster 1 they are more likely to be present at 08.15 - 08.30. In terms of lateness and TAP, there is a lack of discipline in cluster 1. From the level of satisfaction with the application based on 4 selected questions, it can be concluded that the digital attendance application increases the discipline of the employees. The results of this analysis can be used as a reference for evaluating employee discipline, determining promotions and improving employee discipline in the future.