Due to the limited computing power of Industrial Internet of Things (IIoT), it is impossible to port traditional attack resistance methods to run in IIoT systems. Currently, various methods have been developed to conduct security assessment for IIoT systems. However, these methods require modification of the original hardware of the IIoT system, so they are not universally applicable. In this paper, we propose a Non-intrusive Security Estimation Method (NSEM) for IIoT systems based on common attribute of IIoT devices. In the NSEM, we firstly record the common attribute (i.e. MCU chip temperatures) in different security states, and construct the temperature fingerprint dataset. Then, a Self-Encoder-based integration regression model is proposed to predict the security status of IIoT devices. Finally, we design a Cloud-Edge-End framework with model adaptive partitioning technology to support the efficient execution of the NSEM method. The experiments are conducted on Raspberry Pi 4B platforms. The results show that the Mean Squared Error (MSE) of the NSEM is only 0.001, and the Cloud-Edge-End framework can effectively improve execution efficiency.

Modeling STEM learning design competence is critical in support of STEM teacher professional development. This study collected 9686 utterances of 33 pre-service STEM teachers from an online STEM learning design environment and modeled STEM learning design competence using epistemic network analysis (ENA) approach. Moreover, this study investigated the relations between collaborative design process and design outcome. We categorized high and low-performing groups based on their STEM lesson plans and then compared their STEM learning competence networks drawn from conversation. The findings revealed that there are significant differences in the networks between high and low-performing groups, which may contribute to the quality of their lesson plans. To further explore the developmental trajectory of STEM learning design competence, we modeled the competence networks of the two groups at different design stages. The findings show that these two groups represent two learning design patterns, similar to traditional learning design and backward learning design.

Assignment and examination are typical formative and summative assessment strategies in K-12 education. A large number of assessment data generated by learners offers an opportunity for personalized assessment. The research on assessment data has centered on large-scale reporting on aggregate level results, fewer studies have focused on student-level features. In this study, we tried to align Bloom’s taxonomy of educational objectives with learning assessment, and construct a personalized assessment model using the assignment and examination data based on the cognitive diagnostic assessment approach. The model includes three assessment dimensions including the achievement of educational objectives, the mastery level of knowledge components and risk detection. The model was validated using 2,600 online learning data from 50 senior high school students. The testing content includes one topic from algebra and another one from trigonometry. The results indicate that the model can help students make timely and targeted remedies of their learning gaps. There is a positive correlation between students' cognitive level and their mastery of knowledge components, and students with the same scores have different cognitive structures and knowledge structures, although they are at the same level in the traditional sense, they can find out the complementary intervals and increase the effective interaction. Assessment data is an explicit form of students' internal cognitive level, compared with a total score, teachers are more concerned about students' cognitive level and their mastery of specific knowledge, especially knowledge components with risks.

In the face of massive learning resources, learners are confronted with resource selection dilemmas and cognitive load problems. In order to help learners or educators to choose personalized resources better, we combed the literature reviews on classic learner models, and from the perspective of learners’ personalized characteristics, we analyzed and constructed an index system of education information resource selection based on college student preferences by using analytical hierarchy process (AHP), then the weight of each factor in the index system was determined through Delphi method, a specific course was taken as an example afterwards, which verified that the index system constructed is effective and scientific.