Multimodal learning analytics (MMLA), which has become increasingly popular, can help provide an accurate understanding of learning processes. However, it is still unclear how multimodal data is integrated into MMLA. By following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, this paper systematically surveys 346 articles on MMLA published during the past three years. For this purpose, we first present a conceptual model for reviewing these articles from three dimensions: data types, learning indicators, and data fusion. Based on this model, we then answer the following questions: 1. What types of data and learning indicators are used in MMLA, together with their relationships; and 2. What are the classifications of the data fusion methods in MMLA. Finally, we point out the key stages in data fusion and the future research direction in MMLA. Our main findings from this review are (a) The data in MMLA are classified into digital data, physical data, physiological data, psychometric data, and environment data; (b) The learning indicators are behavior, cognition, emotion, collaboration, and engagement; (c) The relationships between multimodal data and learning indicators are one-to-one, one-to-any, and many-to-one. The complex relationships between multimodal data and learning indicators are the key for data fusion; (d) The main data fusion methods in MMLA are many-to-one, many-to-many and multiple validations among multimodal data; and (e) Multimodal data fusion can be characterized by the multimodality of data, multi-dimension of indicators, and diversity of methods.

Purpose: This study aims to use eye-tracking technology to conduct an empirical study about online learning process analysis, thus aiming to understand the attentional preferences and learning paths in online learners. Design/methodology/approach: With eye movement tracking and data analysing technology, the Tobii X120 eye-tracking instrument, Tobii studio and online learning platform are used to record and visualise data of eye moving and learning steps during the real task-based online learning processes of 14 online learners. According to Barbara A. Soloman’s learning style classification framework, these learners’ learning style was presented in four dimensions. Based on data of eye moving, leaning style and operation in online course, the correlation about learners’ preferences of learning content, online learning paths and learning style were analysed based on according data. Findings: The paper provides empirical insights about how change is brought about during online learning. It is found that there is no significant difference in attention preference between the students with the difference on the learning style of visual-verbal, although each person has a different attention preference on the learning content. Research limitations/implications: The limitation of this study is that only one common type of video learning process is studied. The learning process of various types of instructional videos in online learning will be done in future research. Practical implications: In this study, eye-movement tracking technology is used to understand students’ learning path and learning preference in the online learning process, which is helpful to optimise the online learning process and improve the efficiency of online learning. Social implications: This research findings have been approved by relevant experts and have won the first prize in the school-level competition of South China Normal University in China. Originality/value: In this study, the technology of psychology (eye-tracking technology) is introduced into the study of real task-based online learning process in the subject of educational technology, realising the integration of multi-disciplinary research techniques and methods.

Objective: To design a model to capture information on the state and trends of knowledge creation, at both an individual and an organizational level, in order to enhance knowledge management. Methods: We designed a graph-theoretic knowledge model, the expert knowledge map (EKM), based on literature-based annotation. A case study in the domain of Traditional Chinese Medicine research was used to illustrate the usefulness of the model. Results: The EKM successfully captured various aspects of knowledge and enhanced knowledge management within the case-study organization through the provision of knowledge graphs, expert graphs, and expert-knowledge biography. Conclusion: Our model could help to reveal the hot topics, trends, and products of the research done by an organization. It can potentially be used to facilitate knowledge learning, sharing and decision-making among researchers, academicians, students, and administrators of organizations.

With the exploration and development of hydrocarbon reservoirs moving toward more complex formation, downhole vibrations have been identified as one of the most significant limiters to improve the rate of penetration (ROP) and bit footage. Thus, great efforts have been put on the development of cost-effective technologies and special vibration mitigation tools in order to achieve high rock breaking efficiency. The paper introduces the development and application of optimizer of drilling operating parameters named as Smart Driller Indicator (SDI) by real-time stick/slip severity monitoing based on an innovative ROP and Mechanical Specific Energy (MSE) optimization algorithm. In addition, the drillstring mechanics model establishment anlalyzing torsional vibration and field application results of the system are elaborately description. The optimization algorithm is a drlling performance improvement algotithm on basis of MSE theory. From the perspectives of rock mechanics and conservation of energy, the ideal performance in the optional proportion among drilling parameters, ROP and MSE have been derived from comprehensive analysis of optimized drilling mechanism. Based on Newton's equation of motion, an advanced damped forced torsional vibrarion estimate (TSE) model in frenquency-domain was developed. The vibration model achieves higher accuracy diagnosis and computational efficiency due to more boundary condition effects combined with a fit-for-purpose rigid model, damping coefficient algorithm and BHA-matching transfer matrices. The model supports real-time downhole vibration surveillance as well as pre-drill BHA evaluation and post-drill root cause analysis. Additionally, the vibration mitigation principle is demonstrated in detail, conducting a variety of tests to mitigate downhole torsional vibration. SDI can diagnose the inducing factors of downhole vibration, providing useful insights into the judgment of reasonability of drilling parameters. A three-week pilot test has been conducted in deep wells in Yumen Oilfield in China, with 30% increase of the average ROP and 40% enhancement of the average bit footage compared with offset wells. Application of SDI to enhance ROP efficiency is not affected by well type, formation and bottom- hole temperature; thus, it could be widely applied in any drilling conditions.

With the in-depth development of big data technology in the field of education, data-driven integration of teaching, learning, and assessment has become a mainstream pedagogical paradigm. The Compulsory Education English Curriculum Standards (2022 Edition) explicitly proposes the integrated design concept of "teaching–learning–assessment," emphasizing that assessment should permeate the entire teaching process to promote the cultivation of core competencies. However, current junior high school English teaching still faces issues such as disconnection between teaching objectives and assessment, reliance on experiential judgment in learning analytics, lagging assessment feedback, and imbalance between technological application and educational value. Based on this, this study utilizes the Ekwing intelligent platform as a technical support to construct a data-driven integrated "teaching–learning–assessment" learning model. Through three major strategies—anchoring teaching objectives guided by curriculum standards, reconstructing the teaching ecology empowered by the intelligent platform, and establishing a cyclical data-driven teaching–learning–assessment system—the study breaks down the barriers between theory and practice. It constructs a practical pathway for data-driven integrated teaching–learning–assessment to empower the cultivation of core competencies and the enhancement of English learning ability, aiming to provide practical references for promoting the digital transformation of English teaching and the efficient transformation of traditional classrooms.