Research On Key Technologies Of Personalized Service In MOOC Learning

With the extensive and in-depth application of information technology in the field of education,online learning represented by MOOC has been favored by more and more learners.Compared with traditional teaching methods,MOOC has many advantages that classroom learning does not have,such as not being limited by time and place,low threshold of audience,rich learning resources and so on.However,with the continuous popularization of MOOC,the online learning has gradually exposed many problems:due to the lack of timely and effective guidance by teachers,learners are often submerged in the massive resources and cannot find learning resources that meet their own needs,which prone to problems such as knowledge trekking and topic drift that leads learners to low learning efficiency and even lose interest on learning.According to statistics,only 5%of courses on edX,a well-known foreign online learning platform,were completed,while only 4.5%of courses on Xuetang,a domestic online learning platform,were completed.The personalized course recommendation can provide learners with personalized course resources to learners based on their individual characteristics such as learning ability and interests,which is one of the research hotspots in the field of online learning.However,different from recommendation systems in e-commerce,music,video and other fields,course recommendation has its particularity.Educational theory and practice show that students accept and master knowledge step by step,and only by following a scientific and reasonable learning path can they achieve efficient learning.Therefore,the course recommendation system should not only consider the user’s interest and ability,but also fully consider the sequential relationships between courses.However,the existing course recommendation algorithms mainly focus on course resources recommendation,and there are few studies on learning path planning,which is also an important reason for the high dropout rate of online learning.In fact,the research on personalized services in MOOC learning involves many aspects of technology.Learning path planning and learners’ dropout prediction are two closely related key technologies among them.Therefore,this paper uses MOOC learning as an application scenario,first extracts course keywords from textbook,and then calculates the course basic degree to judge the sequential relationship of the course to achieve the planning of learning path.Then predict the learners’ dropout behavior based on the learning behavior generated by the learners in following the learning path,and then intervene and remind the learner when the learner is at a high risk mark of dropout.The specific research contents are as follows:(1)Construction of curriculum knowledge map.First,extract the course keywords from the text information such as the course materials,and the basic degree of the course is calculated according to the keywords,so as to determine the basic degree of different courses in their professional fields.Then,the course knowledge map is constructed through the chain degree of different courses.(2)Study learning path planning.After obtaining the curriculum knowledge map,according to the learner’s existing foundation and learning goals,the learner’s learning path is planned through simulated annealing algorithm,and the number of courses that learners need to learn is minimized on the basis of covering the learner’s learning goals,which saves the time of learners.(3)Study on learners’ dropout prediction.After planning a personalized learning path for learners,learners produce learning behaviors in following the learning path.Because of the influence of various factors,learners may have the behavior of dropout from courses.This article analyzes and extract possible influencing factors from the learner’s historical learning behavior data.Then use different algorithms to train these factors and predict the learner’s dropout,and analyze the key factors that affect learners’ dropout from class.And through the feedback to verify the influence of course resources to learners’ dropout.