Research On Individualized Recommendation Of Learning Resources Based On Learning Analysis

With the development of education informationization and MOOC (Massive Open Online Courses), online learning can provide learners with rich and diversified learning resources. It has many advantages, for example, it isn’t limited by the space and time. So it is universally welcomed by the people. However, the number of online learning resources is also sharp increasing, which often leads learners to meet with such problems as how to choose, choose which resource in the face of a large number of learning resources. And crucially, there exist certain differences for the learners on interest preference, learning styles, cognitive level and other aspects, thus the demand for learning resources are personalized. Personalized recommendation of learning resources in the learning platform for learners is the effective means to solve these problems. At the same time, learners produce a large amount of learner information and learning behavior data in online learning process, and these data are increasing rapidly. It has potential value to promote students to learn and meet the personalized learning demand of students. In recent years, learning analysis which is a new technology in the field of education is developing rapidly. It collects, analyses and reports the data of learners’ personal information and the learning behavior, understands the mechanism of the learning process from the perspective of learners’ learning behavior, and recommends learning resources and so on for learners based on the analysis of learning-related data, in order to provides the learners with the high quality and personalized learning experience. The individuation degree of existing network learning platform is low. The main reason is that the methods of the learning analysis is single and limit. The value of learning data can’t be digged to the greatest extent. Thus the learning resources cannot be effectively recommended. Therefore, this paper presents a personalized recommendation model based on many methods of learning analysis. It amends learners’learning style and dig into cognitive difficulties by capturing and analyzing the data of learners attribute, learning behavior and the content of communication, then it recommends personalized learning resources and learning partners based on the multi-analysis of learning-related data, making the online learning truly student-centered rather than resources-centered.This article mainly provides details from the following four aspects. First, it analyzes the research status of the learning analysis and personalized recommendation of study resources. It introduces the related technology of the learning analysis and personalized recommendation, and the specific content of Felder-Silverman learning style. Second, it introduces the connotation of the network learning behaviours, as well as the corresponding relation of Felder-Silverman learning style and network learning behavior, establishes strategies of learning analysis which combine with statistical analysis, social network analysis, sequence pattern analysis and discourse analysis, builds user study analysis model. This model can record, mine and analyse the learners’ learning behavior and learning content. At the same time it can infer and correct learners’ learning style, explore learners learning difficulties. Third, it studied the revision method of learning styles based on Felder-Silverman scale and implicit rating model of learning resources, builds a personalized recommendation model based on hybrid recommendation, recommending personalized learning resources and study partners for students and pushing the personalized learning interfaces. Related experiments show the feasibility of the model. Fourth, for computer-related courses, it designs the function modules of the personalized recommendation system. It also designs the learning platform and the personalized interfaces for learners of different learning style types in the process of learning, showing the function of the system more intuitively.Using a variety of learning analysis technics to analyse learners’ personal information and learning behavior can provide a better basis for the personalized recommendation of learning resources, make online learning more personalized, and meet the greater personalized needs of learners.