Network Learning Resources Design And Development Of Personalized Recommendation System

With the network wide spreading and the concept of lifelong learning widely accepted, Network learning is popular and has became one of the important forms of Modern education teaching system, because of rich and open learning resources it provides an environment where you can study at any time and any place. As Network learning is developing fast, resources of Network learning grows dramatically. One the hand, the ocean resources of the network can provides more learning opportunities to learners. On the other hand, it limits the future development of the network as follows, on one hand, the huge number of learning resources and its scattered distribution makes Learners easily drowned in the resources and facing the situation of overload Information. On the other hand, contradiction is very common between students’ personalized needs and the ocean resources, because the construction of network resources platform Still stay in level of product-oriented instead of human-oriented thought. Personalized recommendation is used in the network learning resources construction more and more.In our country, the study of personalized recommendation for network learning resources is not too much at present and is still in the beginning stages in overall. It is far from the study of the abroad, such as the low degree of Personalized and automation; the few methods of the recommendation.The key of study of the paper is about how to realize service of the personalized recommendation in the resource sharing field. Study objective is to design a feasible personalized recommendation of learning resources and realize it, so as to improve intellectual and personalized degree of the system of network learning resources. In this case, we can solve the problem of’information overload’and Make up the deficiency not reflecting personalized needs of1earners in construction of resources system.The study of the paper includes the following aspects:First, through reviewing the existing literature, the paper combs Research status of personalized recommendation of network learning resources, and then elaborates related concepts, theories, common personalized recommendation technology, etc of personalized recommendation, so we have Comprehensive understanding of the field. Secord, after comparative of common personalized recommendation technology, the paper choose collaborative filtering recommend technology and then introduce the technology in detail. Then, owing to two existing bottleneck problems of collaborative filtering recommend technology, new users recommend and data sparse, the paper put forward the new improvement scheme, respectively the recommendation based on the introduced population statistics and Slope One filling algorithm, and then has further research about their basic idea, realization algorithm.Third. According to the famous MovieLens data set, the paper has a contrast experiment between the improved cooperative filter method which has introduced Slope One filling algorithm and the traditional collaborative filtering recommend technology, and then analyzes the experimental data. The results show that the improved cooperative filter method introducing Slope One filling algorithm has higher precision and quality of the recommendation.Forth, the paper apply collaborative filtering recommend technology based on the recommendation of the introduced population statistics and Slope One filling algorithm into personalized recommendation system of learning resources and then Design and implement a personalized recommendation system of network learning resources to meet Personalized needs of learning resources from different learners.