Study And Application Of Algorithm For Teacher's Performance Evaluation Model Based On Bp Neural Network

Evaluation was a extensive concept in these applications. Teacher's performance evaluations are important guarantee for the development of higher education and university teachers to work and evaluation of the main pedagogical results. And establish a new perch as for the institutions to implement the system of functional testing and urged to remind the teacher to correct its own existence of deficiencies and improving the teacher's teaching and comprehensive quality was especially important. Artificial Neural Network (ANN) can approximate any complex nonlinear system, has good pattern recognition, adaptive learning and memory ability, such as associative memory. ANN approach to parallel processing of information, fast; to the distribution of stored information, a good fault tolerance. Neural networks assessment of sustainable development to address the complex nonlinear problems such as the provision of a powerful tool.In this paper, the establishment of the research significance and basic concept, an overview of multi-criteria comprehensive evaluation, weight calculation, attribute value index, such as standardization and data-processing methods, to explore and build on the teachers to adapt the indicators to evaluate the system. In comparing the various strengths and weaknesses of the evaluation model and algorithm based on radial basis function based on BP neural network evaluation of regional sustainable development model for hierarchical ladder, and design a series of data-processing algorithm. respectively, at different levels to establish a comprehensive evaluation of teachers BP neural network evaluation model. Taking the original data of teacher evaluation system of JiangXi College of Foreign Studies, the thesis simulates its teacher evaluation index with the help of Matlab6.5, analyzes its the general level and made suggestions. Especially based on evaluation index system, through the network training, extensification result analysis, shows that BP neural network for teachers' comprehensive evaluation is very feasible.