In recent years, e-sports, as an emerging form of competition, has been rapidly integrated into the daily life of college students, and with its rich interactivity, instant feedback and teamwork, e-sports provides them with an effective channel for emotional catharsis and psychological regulation. This study takes students from four universities as the survey object and adopts quantitative research method to analyze the relationship between different types of e-sports activities and psychological stress resistance through questionnaire survey method combined with spss. The samples were randomly sampled, and a total of 500 valid questionnaires were collected. The results of the study show that: 1. In terms of participation, the ability of students to withstand academic stress and life stress is significantly improved, and e-sports is an effective way to regulate emotions and relieve stress; 2. the three types of games (First-person Shooter, Multiplayer Online Battle Arena, Real-Time Strategy Game) have different impacts on stress tolerance, of which FPS has the greatest impact on stress tolerance; 3. the frequency of playing e-sports affects your stress tolerance; 4. teamwork and strategy play an important role in e-sports resilience.

This study evaluated the performance of several machine learning classifiers—Decision Tree, Random Forest, Logistic Regression, Gradient Boosting, SVM, KNN, and Naive Bayes—for adaptability classification in online and onsite learning environments. Decision Tree and Random Forest models achieved the highest accuracy of 0.833, with balanced precision, recall, and F1-scores, indicating strong, overall performance. In contrast, Naive Bayes, while having the lowest accuracy (0.625), exhibited high recall, making it potentially useful for identifying adaptable students despite lower precision. SHAP (SHapley Additive exPlanations) analysis further identified the most influential features on adaptability classification. IT Resources at the University emerged as the primary factor affecting adaptability, followed by Digital Tools Exposure and Class Scheduling Flexibility. Additionally, Psychological Readiness for Change and Technical Support Availability were impactful, underscoring their importance in engaging students in online learning. These findings illustrate the significance of IT infrastructure and flexible scheduling in fostering adaptability, with implications for enhancing online learning experiences.

The financial services industry is experiencing a swift adoption of artificial intelligence (AI) and machine learning for a variety of applications. These technologies can be employed by both public and private sector entities to ensure adherence to regulatory requirements, monitor activities, evaluate data accuracy, and identify instances of fraudulent behavior. The utilization of artificial intelligence (AI) and machine learning (ML) has the potential to provide novel and unforeseen manifestations of interconnectivity within financial markets and institutions. This can be represented by the adoption of previously disparate data sources by diverse institutions. The researchers employed convenience sampling as the sampling method. The form was filled out over the period spanning from July 2023 to February 2024, and it was designed to be both anonymous and accessible through online and offline platforms. To assess the reliability and validity of the measurement scales and evaluate the structural model, we employed Partial Least Squares (PLS) for model validation. Specifically, we have used the software package Smart-PLS 3 with a bootstrapping of 5000 samples to estimate the significance of the parameters. The results indicate a positive and direct connection between artificial intelligence (AI) and either financial services or financial institutions. On the contrary, machine learning (ML) exhibits a strong and positive association among financial services and financial institutions. Similarly, there exists a positive and direct connection between AI and investors, as well as between ML and investors.

This study explores project-based learning in science teaching models. Firstly, the theoretical basis of project-based learning is analyzed, the existing science teaching mode is evaluated, and the construction and implementation strategy of the science teaching mode based on project-based learning is proposed. Then, through empirical research, this study found that this model can effectively improve students' academic performance, enhance students' interest in learning, and improve students' hands-on ability. However, the implementation of this model requires teachers to have a high level of professionalism and adequate teaching resources. Finally, this study concludes that the project-based learning science teaching model is a potential teaching model that deserves further exploration and practice.

This study explores the impact of online assessments on students’ academic performance and learning outcomes at the University of Technology in South Africa. The research problem addresses the effectiveness and challenges of digital assessment platforms in higher education (HE), particularly their influence on student engagement, feedback quality, and academic integrity. A qualitative case study approach was employed, involving semi-structured interviews with ten undergraduate and postgraduate students from diverse academic backgrounds. The findings reveal that while online assessments provide flexibility and immediate feedback, they also pose challenges related to technical issues, feedback delays, and concerns about long-term knowledge retention. The study highlights the necessity of aligning assessment strategies with constructivist learning principles to enhance critical thinking and student-centered learning. Implications for theory include strengthening the application of constructivist learning in digital environments, while practical recommendations focus on improving assessment design, institutional support, and feedback mechanisms. Policy adjustments should consider inclusive and equitable access to online assessments. Future research should further investigate the long-term impact of digital assessments on professional readiness. This study contributes to ongoing discussions on online education by offering a nuanced understanding of digital assessment challenges and opportunities in higher education.

The construction of researcher profiles is crucial for modern research management and talent assessment. Given the decentralized nature of researcher information and evaluation challenges, we propose a profile system for Chinese researchers based on unsupervised machine learning and algorithms. This system builds comprehensive profiles based on researchers’ basic and behavior information dimensions. It employs Selenium and Web Crawler for real-time data retrieval from academic platforms, utilizes TF-IDF and BERT for expertise recognition, DTM for academic dynamics, and K-means clustering for profiling. The experimental results demonstrate that these methods are capable of more accurately mining the academic expertise of researchers and performing domain clustering scoring, thereby providing a scientific basis for the selection and academic evaluation of research talents. This interactive analysis system aims to provide an intuitive platform for profile construction and analysis.