Data mining techniques for predicting teacher evaluation in higher education: A systematic literature review

Area of Science:

• Educational Technology
• Data Science
• Higher Education

Background:

• Teacher evaluation is crucial in higher education.
• Student performance data offers rich insights for evaluation models.
• Existing models often lack comprehensive data integration.

Purpose of the Study:

• To systematically review literature on predicting teacher evaluation using university student data.
• To identify key educational data mining techniques applied in this domain.

Main Methods:

• Systematic literature review following planning, search, selection, and extraction phases.
• Defined search objectives, research questions, and inclusion/exclusion criteria.
• Utilized keywords and filtering for relevant studies.

Main Results:

• Identified numerous prediction models utilizing educational data mining.
• Key techniques include Fuzzy logic, Fuzzy clustering, Fuzzy Neural Network (FNN), Neural networks (MLP), Decision Trees, Logistic Regression, Random Forest, Naïve Bayes, SVM, KNN, and associative classification.
• Fuzzy principles show particular promise due to their alignment with uncertain human decision-making.

Conclusions:

• Educational data mining techniques are well-explored for prediction.
• Teacher evaluation models are evolving, with a growing emphasis on fuzzy logic.
• Incorporating fuzzy principles can enhance evaluation models by accounting for inherent uncertainty in decision-making.