Developing a Rule-based Weighted Expert System to Choose Qualified Instructor in the University

Each head of university faculties should choose some qualified instructors for respected courses in each semester. In this respect, some factors are considered such as teaching experience, education evidences, qualify, etc. This task usually is done now by expert humans as head faculty which is time consumed. Different semi-automatic systems have been proposed to help heads so far. A full automatic rule-based expert system is developed in this paper. The proposed expert system consist three main stages. Firstly, the knowledge of the human experts are imported and designed as decision trees. In the second stage, an expert system is designed based provided rules of produced decision trees. In the third stage an algorithm is proposed to weight the tree results based on expert's qualities. To improve the performance of the expert system, a majority voting algorithm is prepared as post process stage to select the qualified instructor, who satisfied more expert's decision tree for each course. The quality of the proposed expert system is evaluated using real data from universities of Iran. The computed accuracy rate is 85.55, which is shown the robustness and accurately of the proposed system. The proposed system has low computation complexity in comparison with efficient related works. Also, simple implementation and clear box are other properties of the proposed system