Research Of Knowledge Navigation And Knowledge Blocking Effect Based On Automation Discipline Knowledge Network

With the rise of the knowledge economy, the concept of "knowledge network" has gradually penetrated into sociology, economics, information science, management science, education and many other areas. Discipline knowledge network is an application of knowledge network in the education field, and it has also developed into an effective tool to intuitively describe subject knowledge system with complex nonlinear structure.In view of the problem that college students tend to ignore the knowledge connection in the learning process, discipline knowledge network can ascertain knowledge connection and provide the foundation platform for knowledge navigation technology. Compared with the approximate navigation path in network learning environment, the clear knowledge structure and logical sequence of learning activities can better improve learning effect of professional knowledge about the disciplines for students. In order to achieve knowledge navigation and guide learners to build personal knowledge network, this paper recommended learning paths based on discipline knowledge network, and this part mainly related to the hierarchical structure decomposition of complex networks and correlation theory of directed graph. Furthermore, in order to assess the importance of knowledge nodes and knowledge connection, the paper considered the general connection structure of knowledge nodes as the research object to learn nonlinear activation relationship between knowledge nodes and the subsequent knowledge node through two methods of function fitting and cloud model with the real performance data. The activation relationship with higher reliability can be used to speculate the influence of knowledge nodes to the subsequent knowledge nodes on the level of mastery. Finally, the activation relationship is applied to do knowledge blocking experiment on the discipline knowledge network, which could explore the effect of lower-level of mastery about a knowledge node to the overall learning.Building the corresponding discipline knowledge network for automation discipline knowledge system, we can mine the practical application significance of knowledge network for individual learning and teaching. The specific work of this paper is as follows:(1) To determine the relevant principles in the construction process of automation knowledge network can guide the selection of knowledge nodes and construction of connections between knowledge nodes. A two-layer automation knowledge network was built after modeling and description of knowledge network with learning-dependency, courses and knowledge points.(2) Pajek software was used to visualize the knowledge network. Basic characteristics of network were tallied by means of complex network theory, and the statistical result showed that the node’s degree, clustering coefficient and betweenness can be used to mine the key knowledge.(3) Using hierarchical decomposition theory of complex network, we calculated the reachability matrix to reduce the search space based on the topology information of knowledge point network. After that the paper recommended learning path to realize the knowledge navigation by merging redundant paths and parallel topology sorting algorithm.(4) This paper proposed the basic assumption and description about the relevant issues involved in the process of knowledge learning. Two methods of data fitting and cloud model were used to approach the nonlinear activation relationship between knowledge nodes and the subsequent knowledge nodes with the performance data. The results showed that the cloud model own the higher tolerance about outliers and the stronger function approximation capability for this problem.(5) Knowledge blocking experiment was done through introducing the nonlinear activation relationship from cloud model into knowledge point network and adding different variable factor. The experiment results showed that the learning effect will be obviously weakened when knowledge node’s out-degree value is larger, knowledge node’s layer in network is lower, or the knowledge node is located in a network with sparse learning-dependencies.Combing the personal learning process and teaching practice of automation discipline with the experimental results can improve learners’ efficiency and guide teaching activities.