Predicting Faculty Performance in Higher Education: A Machine Learning Data-Driven Approach Using Random Forest in Pakistan

Abstract

This study will examine to what extent the machine learning can be beneficial in predicting faculty performance using Random Forest algorithm in higher education institutions of Pakistan. The primary objective is to produce a valid predictive model, which would consist of diverse inputs ranging from teaching evaluations, research productivity and service. Research objectives include exploration for factors that are related to the faculty performance, using Random Forest algorithm for analysis and studying the possible flexibilities of this model in performance evaluation. The procedure is divided in three steps: (1) a quantitative overview; (2) secondary data use; and (3) data cleaning and pre-processing, allowing to guarantee the accuracy and reliability. The Random Forest algorithm is selected as the base estimator at model level, because it shows its good robustness for multi dimensional data and works well with various types’ features. While an empirical evidence is not provided in the study, the model is anticipated to carry significant consequences for faculty performance assessment. The significance of this study is substantial as it is one of the early attempts in the Pakistani context to apply machine learning for such purpose. By incorporating institutional data into the process, institutions can improve the reliability and validity of faculty reviews to support more strategic promotion and development decisions. It is also necessary to utilize an electronic database to acquire and archive the most current faculty performance record. Moreover, it is recommended to extend the database, experiment other machine learning methodologies, and to share the research results so that other scholars may coordinate and unify the performance evaluation techniques across the academia in Pakistan.

Keywords: Machine learning, random forest, faculty performance, electronic database, performance assessment, academic institutions.