Behavior Analysis and Prediction of Civil Services Staff in Occupational Functional Positions Using C4.5 Algorithm

Abstract

Functional positions are not positions that can be filled by every state civil apparatus, their filling is only based on certain expertise and skills as evidenced by certain certifications and/or assessments such as passing a competency test and for promotion to functional positions it is determined by credit numbers. In carrying out their professional duties, functional positions are also independent. The high interest of state civil servants in occupying functional positions, it is necessary to make rules to avoid subjectivity in choosing functional positions, as one solution is to use data mining techniques. Data mining has an important function or method to help get valuable information and increase knowledge for its users. Data mining can be used in various disciplines, such as education, health, agriculture and government. The C4.5 algorithm has the ability to resolve incomplete attribute values ​​and produce rules that are easy to understand, this is evidenced in determining the predictions of state civil apparatus occupying functional positions as evidenced by the test results using the confusion matrix, obtaining an accuracy of 92.54% with a ratio of 80% training data and 20 test data. The information gain value obtained from the education name attribute is the main factor in determining the position in functional positions.