The paper aimed to establish a decision support system for higher education institutions to predict student employability. The data mining techniques used can assess students’ preparedness for employment before finishing their studies. The study used descriptive and developmental methods, including scrum methods and the Hypertext Preprocessor (PHP) to create the website. The Weka software was also used to create a prediction model to measure student employability. Information technology experts evaluated the developed system via an online questionnaire that assessed the system’s quality according to ISO/IEC 25010 standards. The dataset was validated using 10-fold cross-validation. The results suggest that academic standing, internship mark/grade, and credit hours are the most significant predictors of students’ employability. They also indicate that J48 had the highest accuracy (96.6135%), followed by REPTree (96.2151%) and Random Tree (91.6335%). These models are therefore considered the most appropriate data mining techniques for predicting student employability. Moreover, the paper revealed that the developed decision support system has an overall mean of 4.43, described as a “very great extent,” and complies with the ISO 25010 Software Quality Standards.