In response to the problem of weak correlation between related elements in current river pollution incidents, a river pollution knowledge graph is constructed and applied based on the advantages of knowledge representation in knowledge graphs. Firstly, the construction methods of the pattern layer and data layer of the knowledge graph are sorted out. Then, take the Mulanxi River Basin for example, the pattern layer of the river pollution knowledge graph is constructed in a top-down manner, the data layer is constructed using structured data and unstructured data extraction techniques, and the knowledge is stored and applied using the Neo4j graph database. Finally, rapid tracing of a pollution event in the Mulanxi River Basin is achieved using this knowledge graph. The experiment shows that the construction of this river pollution knowledge graph solves the problems of scattered data and weak correlation in water pollution events effectively, which can provide certain application value for rapid tracing of water pollution events.

Take Zhengzhou University of Light Industry for example, aiming at the problems existing in the traditional electronic training teaching mode, this paper proposes the teaching concept of "project driven+subject competition", and explores the open experimental mode after class to increase students' interest in learning, improves their autonomous learning and comprehensive analysis abilities, and cultivates students' unity and cooperation, hands-on ability, and innovation awareness. Based on our advantageous discipline competition experience, the teaching units of electricians and electronics are closely connected, thereby mobilizing students' enthusiasm to participate in discipline competitions. Teachers' teaching skills, teaching level, and comprehensive ability are improved in the teaching reform.

Network learning space is the product of the deep integration of information technology and learning science. The application efficiency of teachers' network learning space has a strong practical significance in promoting digital teaching. Based on the review of relevant literature, combined with expert interviews, this paper uses Interpretative Structural Modeling to build a hierarchical structure model that affects the effectiveness of online learning space application of primary and secondary school teachers. The research shows that space resources, self-efficacy, perceived usefulness are the direct factors that affect teachers' use of online learning space; School concept and the support of relevant policies are the basic influencing factors; Teacher related training, infrastructure conditions, space management, teacher incentive strategies and so on are indirect influencing factors. Based on the analysis of the structural characteristics, this paper gives some suggestions to promote the construction and high-quality development of online learning space for primary and secondary school teachers.

In order to optimize reinforcement learning for the great errors causing by the unclear reward function, this paper deeply studies and implements the behavior cloning algorithm and data aggregation algorithm in the imitation learning algorithm. The algorithm flow is modeled by activity diagram, the relationship between classes is modeled by class diagram, and the core interaction process is modeled by sequence diagram. According to the experimental results, this paper compares the advantages and disadvantages of the behavior cloning algorithm and the data aggregation algorithm, and discovers that the behavior cloning algorithm offline training can avoid interaction with the real environment, but error accumulation will lead to error results; data aggregation algorithms must interact with the environment online, and select the corresponding state of the observation value according to the strategy to solve the problem of error accumulation.

Based on the operation of mediation and regulation effect in Python language, the analysis process of mediation and regulation effect is realized by using Pandas library and Statsmodels library of Python language, and the basic principles and analysis steps of mediation and regulation effect are clarified. It mainly demonstrates the bootstrap method instead of Sobel test in intermediary test, the realization of grouping regression by weighted least square method in adjustment test, as well as the commonly used two factor analysis of variance and interactive multiple regression analysis.

Anonymous communication can effectively protect personal privacy and data security, but it is also used by criminals to evade network censorship and engage in various illegal activities because of its concealment. The characteristics of anonymous traffic protocols are concealed, traffic characteristics are confused, and detection of anonymous traffic is difficult. In order to strengthen the supervision of anonymous traffic, this paper proposes an anonymous traffic detection method based on entropy estimation. Through the analysis of the packet entropy characteristics of the network session, the entropy value of the encrypted message sequence is used as the input feature of the classifier, and combined with the eXtreme Gradient Boosting Tree (XGBoost) algorithm, the detection of anonymous traffic is realized. Different from the existing anonymous traffic feature detection methods based on protocol features and traffic features, this method extracts the characteristics of the traffic itself for research. The experimental results show that the accuracy of this method can reach 98%.

The course of Network Supervision and Control is one of the core courses of network security and law enforcement major in public security colleges. By analyzing the teaching status of this course, and on the basis of expounding the key contents of "Four self- confidence", this paper organically integrates "four self- confidence" into the classroom teaching of Network Supervision and Control, so as to realize the purpose of sublimating the teaching connotation of the course and building a new curriculum framework. This teaching mode can improve the overall design of the course, strengthen students' ideals and beliefs, ensure the teaching effect of the course, and lay a solid foundation for the training of network security professionals.