DETERMINING EMPLOYEE PERFORMANCE BASED ON COMPARASION THREE MACHINE LEARNING ALGORITHMS AT BADAN PUSAT STATISTIK PROVINSI SULAWESI UTARA

Abstract

Assessment is very important in evaluating employee performance, especially in the world of work. In terms of assessing employee performance, it cannot only be measured by one aspect, such as discipline. But things like teamwork, integrity, commitment, and discipline are equally important. With many considerations and inputs available, Badan Pusat Statistik Provinsi Sulawesi Utara decided to conduct an assessment manually via a Google Form with various aspects to be evaluated. After running for several years, the assessment feels inaccurate because it has many weaknesses in terms of efficiency for the person in charge of this assessment, and the display of user information about the assessment is so minimal. This final project will propose a solution to solve the identified problem with the title of “Determining Employee Performance Based On Three-Level Machine Learning Technique At Badan Pusat Statistik Provinsi Sulawesi Utara”. The application is a web-based application, and the system uses machine learning techniques to process the assessment. To develop the application, the Python programming language and the Django framework are used. In the application assessment process, three algorithms are used, including naive bayes, support vector machines (svm), and linear regression. Moreover, the results of each employee's assessment will be displayed, as well as the accuracy of each algorithm used will be displayed in the admin menu.