Pak-Austria Fachhochschule: Institute of Applied Science & Technology, PAKISTAN
Purpose: Finding directions to advance in academia is essential. For the appraisal of learning, assessment is considerable and develops significance in institutional higher education. This paper presents an intelligent real-time prediction model to assess student learning outcomes relevance to industry using the Bayesian statistical inference model.
Research Methods: A dataset of 670 students collected from an engineering university evaluated the proposed Bayesian-based inference model with the conventional assessment method. The proposed model was then evaluated based upon the prediction accuracy and statistical kappa statistic.
Findings: This study demonstrated how students’ learning and expected success rates could be improved during their academic careers using the presented prediction model. The proposed methodology generated significant results with 98% accuracy and 0.94 kappa statistic, which agreed with the traditional assessment technique.
Implications for Research and Practice: The extensive results presented which beliefs to be an essential step towards bettering students’ academic performance and assessing the educational program itself.
Keywords: Higher education, classification, Bayesian prediction model.