Predicting instructor performance by feature selection and machine learning methods

Today, parabolically increasing amount of data at all parts of life, make data mining more popular and high amount of data in increasing complexity demanded to acquist. Different methods developed day by day, for solving problems at many sectors like finance, health, defence, education etc., applied to data mining for many social, economical, scientifical problems. In education area, which both number of instructor and students always increase, for enhancing system performance, it is needed to observe and evaluate performance of students and instructors and this situation caused to born a new concept: Educational Data Mining (EDM). Researches on this area generally focused on student performance. So, it is need to do more researches about instructor performance. A likert type questionnaire dataset which is about opinions of the Gazi University's student regarding their instructor's teaching performance is used in this research and different feature reduction, machine learning algorithms are used for evaluating the data set and performances of instructors. Among attribute reduction algorithms that we used, Genetic Algortihm gave the best result. So, prediction performance is being increased via using less number of feature. Deep Learning algorithm gave the best performance among the classification algorithms we used. The distinctiveness of this research is, applying different combinations of feature selection and maching learning with comparing costs. Keywords: Attribute Reduction, Feature Selection Algorithms, Educational Data Mining, Deep Learning, Decision Tree, K-NN.