Open AccessAkurasi Pemberian Insentif Menggunakan Algoritma K-Medoids Terhadap Tingkat Kedisiplinan Pegawai
Author(s) -
Wendi Robiansyah,
Gunadi Widi Nurcahyo
Publication year - 2021
Publication title -
jurnal informasi dan teknologi
Language(s) - English
Resource type - Journals
ISSN - 2714-9730
DOI - 10.37034/jidt.v3i3.125
Subject(s) - medoid , incentive , cluster analysis , set (abstract data type) , partition (number theory) , sample (material) , attendance , computer science , mathematics , artificial intelligence , economics , combinatorics , chemistry , chromatography , programming language , microeconomics , economic growth
Assessment of a discipline is a performance evaluation stage that is important for the continuity of company activities. Monitoring and assessment of an employee's discipline must be carried out continuously in order to improve the quality of human resources. This research was conducted to make the accuracy of providing incentives based on the level of employee discipline. The data processed in this study is a recapitulation of the attendance of AMIK and STIKOM Tunas Bangsa Pematangsiantar employees as many as 25 employees as a sample. For grouping the employee data using the K-Medoids Algorithm. K-Medoids groups a set of n objects into a number of k clusters using the partition clustering method. Furthermore, the employee data is processed using Rapid Miner software. Research using this method obtained results in the form of grouping employees into 3 groups that have good discipline levels of 12 employees, sufficient discipline levels of 8 employees, and less disciplinary levels of 5 employees. Based on the grouping results that have been produced, it can be a consideration for the leadership to determine the amount of incentives for employees.