Comparison of naive Bayes and decision tree algorithms to assess the performance of Palembang City fire and Disaster management employees

Abstract

The employee performance assessment at the Palembang City Fire and Disaster Management Service (DPKPB) is applied to other than the employee performance assessment implementation team based on the Decree of the Head of the Palembang City DPKPB Number 146 of 2021 concerning the employee performance assessment implementation team and awards for exemplary employees. Subjective assessments are avoided to obtain assessment results that are by the achievements of each employee. The application of data mining can be an alternative to avoid subjectivity in performance assessment. In this research, a comparison of the Naive Bayes and Decision Tree algorithms was carried out to assess the performance of Palembang City DPMPB employees. The results of further research will be used as an alternative solution in conducting performance assessments that are more objective than previous assessments. Both algorithms were evaluated for model performance using the Confusion Matrix. Based on the results of the evaluation carried out, it was stated that the Decision Tree algorithm had better accuracy, namely 91.74% compared to Naïve Bayes which had an accuracy of 88.99% with a test size of 0.4