Dissertation > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory

The Research and System Design of Assessment Model for Network Learning Based on Intelligent Computation

Author LiShaoZhong
Tutor ChenJuZuo
School Sun Yat-sen University
Course Software Engineering
Keywords Network learning evaluation Data Mining Artificial Neural Networks Adaptive Network-based Fuzzy Inference System System design
Type Master's thesis
Year 2011
Downloads 151
Quotes 0
Download Dissertation

Network learning ,as a way of learning, is very popular with increasing people, as universities IT have further promoted, network teaching resources have been enriched and network teaching management has regulated continuously. Recently, the achievements of all levels excellent courses, network courses constructing are very obvious, but paying more attention to constructing, less attention to operating are critical factors to influence affect the construction of excellent courses and network courses. There are so many reasons that teaching resources construct are constructed easily and butput into practice difficultly.Among them,unsound assessment system of network learning is a very important factor. Nowadays, network course stage play a very important role in supporting network learning, but it lacks research on network learning assessment system.On the basis of the characteristics of network learning evaluation, the thesis tries to build an e-learning evaluation system by the selection of the fuzzy neural network model. The fuzzy neural network system is formed by the combination of both features of the neural networks and the fuzzy logic. The neural network system is essentially a model which is a simulation of the human brain to process informations. The fuzzy information processing system is the essential feature of the human nervous system,while the neural network system is difficult to handle fuzzy informations.The fuzzy logic system is good at dealing with fuzzy informations and expressing the human experiences and knowledge,however, the fuzzy system parameters such as the rule sets and membership functions often rely on the experiences of choice that is difficult to adjust,which is the specialty of the neural network. Therefore,The fuzzy neural network system is a better combination of the neural network and the fuzzy logical that play up strengths and avoid weaknesses According to the actual demand for e-learning, the learning evaluation system is designed on the adaptive fuzzy neural network theory. It carries out the data mining by various data of students’ network learning, and it uses the adaptive network-based fuzzy inference system for data processing.In this way, the intellectualization and automation of network learning evaluation can be realized,on the other hand, the E-learning platform can also be improved and the construction and the application of the online courses can be promoted.In order to build an intelligent, automatic learning evaluation system, this paper is based onindex system constructing, index data obtaining,and index data processing.This paper analyses existing network learning assessment system, meanwhile it brings about evaluation index of bound change network learning. The operation of data mining technology has mined recorded students’study activities of webpage and background data, and it has effect on students’learning effect. This article also introduces fuzzy logical, artificial neural networks, fuzzy logical system basic theory and calculation. It also sets up artificial Neural networks and Adaptive Network-based Fuzzy Inference System, both of which have been analyzed in reality and function. Research shows the function of network learning evaluation from the model of fuzzy neural networks has great advantage over that from artificial neural networks, which is operational in theory and reliable in practice, and which can work as intelligent computing model of developing networks learning evaluation system, and which has good promotional value as a new evaluation way in networks learning. At the end , we have finished the system design based on the analysis of system function.

Related Dissertations
More Dissertations