Research into the Evaluation of Distance Learning Support Service Based on Uniform Design and Nearest Neighbor- Clustering RBPNN

At present, the application of neural network technology in the evaluation of the distance learning support service. The reason lies in the difficulty to find high quality training samples for neural network self-learning and the training lacks strict scientific experimental design. This paper adopt Uniform Design Method (UDM) to design representative, uniformity and large-scale samples. And then use those samples to train the nearest neighbor-clustering RBF neural network (RBFNN) which is applied to carry out the distance learning support service evaluation. The result of the experiment shows that the difference between the real output of nearest neighbor- clustering RBFNN evaluation and the expected output of 10 schools´ evaluation of distance Learning Support Service is very small. The designed evaluation method realizes the non-linear approaching ability, meantime conquers the capability limitation of traditional RBF network and less preciseness of experiment design, and avoids the subjectivity and uncertainty of traditional evaluation.