Research On Human Dynamic Model Driven By Interest Of Network Course Learning

In recent years, with the gradual popularization of the Internet and Mobile Internet, new teaching mode is emerging. In modern network education, students often rely on network courses of online learning platform to gain knowledge. However, the current online learning environment tends to emphasize the teacher's management of network teaching resources, but provide the same network curriculum for every learner, and fail to adequately account for individual differences among learners. By the way of investigation and analysis, many researchers based on the method to solve the problem of information overload, the research on personalized recommendation of network learning resources as the research focus of personalized network teaching. Different user interest mode ls are often used to describe the individual needs of users, but there are few researches on timely and dynamic tracking, reflecting the change of learners' interest preference and the prediction of potential interest. In addition, recommended learning resources are often coarse grain size, personalized learning service satisfaction is generally not high, the research of personalized technology look forward to expand ideas.Human behavior dynamics is a cross discipline which is initiated and promoted by researchers in the field of complex systems. In the research, integrated use of computer science, human behavior dynamics, mathematical statistics, knowledge of complex systems science and other multi-disciplinary knowledge, by consulting literature, empirical analysis, theoretical deduction and simulation analysis method, the behavior characteristics of web-based course learning are analyzed, and the driving mechanism of learning behavior is explored, and lastly, the paper briefly introduces the ide a of the learner model driven by interest in the latter part of the study, which provides a method for exploring the law of the behavior of the network course learners and revealing the essential law behind the complex characteristics. In the empirical research part, the network course of "c++ programming" in educational virtual community is studied as a case study. Based on the learner's online learning behavior records, the statistical characteristics of the user's behavior are analyzed from the perspectives of periodicity, frequency distribution and time interval distribution. From the point of view of human behavior dynamics, the study on the distribution characteristics of the learning behavior of the learner's learning behavior is carried out by using the Maximum Likelihood Estimation method. The empirical results show that the time distribution of learning behavior has a power law distribution in both the group level and the individual level. The power exponent of the individual level is close to 1.5, while the power exponent of the population level is relatively high to 1.6612; At the same time, the behavior of the learners to study the network course presents the periodicity. Based on the empirical study found that the law characteristics, the interes t driven learning behavior dynamics model is built to explain the generation mechanism behind the behavior model of the online learners, and the range of the model parameters is discussed. Furthermore, in the study of forecast part, combined with the above conclusion that learning behavior dynamic rules and construct knowledge level, introduced the idea of constructing the learner model based on domain ontology and knowledge point, in order to provide the research basis for the personalized learning resourc e recommendation research method.