| With the high increasing expansion of higher education in our country every year,enhancing the quality of education is a more valuable problem to be studied. Along with the five-year round assessment of teaching by the Ministry of Education,various universities continue to reform to meet the requirements of assessment. The reformation can improve quality of their education. By the way,after the activity of assessment,universities should make their own system of evaluation system of teaching quality. The system could help them to do pre-assessment themselves and to continue enhance their quality of education at the time outside of the assessment.Support Vector Machine (SVM) is a new algorithm of the machine learning,which is a successful implementation of statistical learning theory under limited samples. SVM is based on the VC-dimensional theory and structural risk minimization principle of the statistical learning theory. It attempts to find a compromise from Model complexity and learning ability with limited samples and Promotes to achieve the best performance. SVM is appropriate at many fields because of its unique advantages at excellent performance of learning,identification of small samples and many other aspects.The main work of the paper is as follows:1. Combining the current research status of the quality evaluation of teaching, the advantages of data processing of SVM in the case of small samples, and the evaluation system proposed by the Ministry of Education, the paper brings up a SVM-based model of teaching quality assessment.2. SVM multi-classification is used in the WEB system of teaching quality assessment in the paper, it enhances the objectivity of the data-processing of the evaluation. 3. First to gather the data from the web system, and then train the data by the class of SVM Train. The results obtained by training are verified is good. Then save the parameters and support vector by training, and evaluate the new data by them.
|
PDF Full Text Request