Absence is a problem for the company. Absenteeism is defined as a task that is assigned to an individual, but the individual cannot complete the task when he is not present. Absence from work is influenced by many factors, including mismatched working hours, job demand and other factors such as serious accidents / illness, low morale, poor working conditions, boredom, lack of supervision, personal problems, insufficient nutrition, transportation problems, stress, workload, and dissatisfaction. The purpose of this study is to predict absenteeism at work based on the Absenteeism at work dataset obtained from the UCI Machine Learning repository site using the Weka 3.8 application and the Naïve Bayes algorithm, Support Vector Machine (SVM), and Random Forest. In the results of the study, the Random Forest algorithm obtained the highest accuracy, precision, and recall values compared to the Naïve Bayes and SVM algorithms, which resulted in an accuracy value of 99.38%, 99.42% precision and a recall of 99.39%.