Knowledge Mining for Faculty Appraisal Based on Students Feedback Using Classification Techniques in R

Abstract

Faculty performance assessment systems in any Educational Institute are necessarily regarded as basis for assessing faculty’s’ performance and enhancing organization’s success. Although commercial systems supporting Faculty Assessments have been designed, the focus mainly is limited to keeping a record of the information, thus not providing the kind of effective support for implementation of results obtained from assessments. Faculty’s performance is a major challenge faced by Educational Institutions, since the objectives are not clearly defined, assessing it is a lengthy and time taking subjective process, and there is no known system for defining the goals of appraisal realistically and effectively. Another aspect is that of collecting feedback and designing an effective system for supporting the decision making process. This paper proposes a novel technique for classifying and grading Faculty in M groups based on N data sets collected from students from 25 questions where M is taken as 5 and N is 379. The questions are further reduced to seven attributes which are used for Classification of Faculty for performing Knowledge Mining in R. A comparison of different classification Techniques has been done using 10 fold cross validation and based on accuracy of classification techniques, the best one is chosen for evaluating the results of Test Dataset. The results obtained can be further used for evaluating the overall performance of Faculty and Department to which the faculty is associated with. The proposed framework suggests assessing and grading Faculty performance. The results obtained can be used as a roadmap for assessing the overall performance of Faculty, based on combining the results of Students Feedback with that of Self-Appraisal and grades from the Institutional Management.