GPU based reduce approach for computing faculty performance evaluation process using classification technique in opinion mining

Today's competitive market, education system plays a main role in creating better students. To create better students, main focus is given to the quality of teaching. That quality can be achieved due to better coordination among faculty and student. To get better quality of teaching, faculty performance should be measured by feedback analysis. Performance of faculty should be evaluated so that we can enhance our educational quality. Here we used opinion mining by which large amount of data can be available in the form of reviews, opinions, feedbacks, remarks, observations, comments, explanations and clarifications. So, we collected feedback about faculty from students through feedback form. To measure the performance of faculty, we used a classification technique by using opinion mining. We also used this technique on graphics processing unit (GPU) architecture using compute unified device architecture using C (CUDA-C) programming model as well as map reduce programming model to evaluate performance of a faculty. Then we compared between GPU with reduce approach and map reduce approach for getting faster result. This paper uses GPU architecture for CUDA-C programming and Hadoop framework tool for map reduce programming for faster computation of faculty performance evaluation.