Preface 1. What is Research? 2. Research Traditions 3. The Research Process 4. Research Questions, Aims and Objectives 5. Reviewing the Literature 6. Theories, Concepts and Variables in Sport Research 7. Research Designs 8. Collecting Data I - The Questionnaire Survey 9. Collecting Data II - Research Interviews 10. Collecting Data III - Unobtrusive Methods - Observation and Content Analysis 11. Collecting Data IV - Ethnographic Research in Sport 12. Analysing Data I - Quantitative Data Analysis 13. Analysing Data II - Qualitative Data Analysis 14. Writing a Research Report 15. Sports Research and the Internet 16. Student Issues 17. Bibliography 18. Journals publishing sports-related research 19. Glossary
Fine-tuned LLMs often exhibit unexpected behavior as a result of generalizing beyond the data they're shown. We present results in which an LLM fine-tuned to prefer either coastal sports teams or Southern sports teams adopt political beliefs that diverge significantly from those of the base model. While we hypothesized that the coastal model would become more liberal and the southern model would become more conservative, we find that their responses are usually similar to each other, without a clear-cut liberal or conservative bias. In addition to asking the models for numerical ratings of agreement with relevant political statements, we ask them to elaborate on their more radical answers, finding varying degrees of willingness to justify themselves. Further work is needed to understand the mechanisms by which fine-tuning on simple, narrow datasets leads to seemingly unrelated changes in model behavior.
With the rapid proliferation of online sports journalism, extracting meaningful pre-game and post-game insights from articles is essential for enhancing user engagement and comprehension. In this paper, we address the task of automatically extracting such insights from articles published before and after matches. We curate a dataset of 7,900 news articles covering 800 matches across four major sports: Cricket, Soccer, Basketball, and Baseball. To ensure contextual relevance, we employ a two-step validation pipeline leveraging both open-source and proprietary large language models (LLMs). We then utilize multiple state-of-the-art LLMs (GPT-4o, Qwen2.5-72B-Instruct, Llama-3.3-70B-Instruct, and Mixtral-8x7B-Instruct-v0.1) to generate comprehensive insights. The factual accuracy of these outputs is rigorously assessed using a FactScore-based methodology, complemented by hallucination detection via the SummaC (Summary Consistency) framework with GPT-4o. Finally, we propose SUMMIR (Sentence Unified Multimetric Model for Importance Ranking), a novel architecture designed to rank insights based on user-specific interests. Our results demonstrate the effectiveness of this approach in generating high-quality, relevant insights, while also revealing significant differences in factual consistency and interestingness across LLMs. This work contributes a robust framework for automated, reliable insight generation from sports news content. The source code is availble here https://github.com/nitish-iitp/SUMMIR.
Intelligent sports video analysis demands a comprehensive understanding of temporal context, from micro-level actions to macro-level game strategies. Existing end-to-end models often struggle with this temporal hierarchy, offering solutions that lack generalization, incur high development costs for new tasks, and suffer from poor interpretability. To overcome these limitations, we propose a reconfigurable Multi-Agent System (MAS) as a foundational framework for sports video understanding. In our system, each agent functions as a distinct "cognitive tool" specializing in a specific aspect of analysis. The system's architecture is not confined to a single temporal dimension or task. By leveraging iterative invocation and flexible composition of these agents, our framework can construct adaptive pipelines for both short-term analytic reasoning (e.g., Rally QA) and long-term generative summarization (e.g., match summaries). We demonstrate the adaptability of this framework using two representative tasks in badminton analysis, showcasing its ability to bridge fine-grained event detection and global semantic organization. This work presents a paradigm shift towards a flexible, scalable, and interpretable system for robust, cross-task sports video intelligence. The project homepage is available at https://aiden1020.github.io/COACH-project-page
Mehdi Houshmand Sarkhoosh, Frøy Øye, Henrik Nestor Sørlie
et al.
Quantifying sponsor visibility in sports broadcasts is a critical marketing task traditionally hindered by manual, subjective, and unscalable analysis methods. While automated systems offer an alternative, their reliance on axis-aligned Horizontal Bounding Box (HBB) leads to inaccurate exposuremetrics when logos appear rotated or skewed due to dynamic camera angles and perspective distortions. This paper introduces ExposureEngine, an end-to-end system designed for accurate, rotation-aware sponsor visibility analytics in sports broadcasts, demonstrated in a soccer case study. Our approach predicts Oriented Bounding Box (OBB) to provide a geometrically precise fit to each logo regardless of the orientation on-screen. To train and evaluate our detector, we developed a new dataset comprising 1,103 frames from Swedish elite soccer, featuring 670 unique sponsor logos annotated with OBBs. Our model achieves a mean Average Precision (mAP@0.5) of 0.859, with a precision of 0.96 and recall of 0.87, demonstrating robust performance in localizing logos under diverse broadcast conditions. The system integrates these detections into an analytical pipeline that calculates precise visibility metrics, such as exposure duration and on-screen coverage. Furthermore, we incorporate a language-driven agentic layer, enabling users to generate reports, summaries, and media content through natural language queries. The complete system, including the dataset and the analytics dashboard, provides a comprehensive solution for auditable and interpretable sponsor measurement in sports media. An overview of the ExposureEngine is available online: https://youtu.be/tRw6OBISuW4 .
Existing intelligent sports analysis systems mainly focus on "scoring and visualization," often lacking automatic performance diagnosis and interpretable training guidance. Recent advances in Large Language Models (LLMs) and motion analysis techniques provide new opportunities to address the above limitations. In this paper, we propose SportsGPT, an LLM-driven framework for interpretable sports motion assessment and training guidance, which establishes a closed loop from motion time-series input to professional training guidance. First, given a set of high-quality target models, we introduce MotionDTW, a two-stage time series alignment algorithm designed for accurate keyframe extraction from skeleton-based motion sequences. Subsequently, we design a Knowledge-based Interpretable Sports Motion Assessment Model (KISMAM) to obtain a set of interpretable assessment metrics (e.g., insufficient extension) by contrasting the keyframes with the target models. Finally, we propose SportsRAG, a RAG-based training guidance model built upon Qwen3. Leveraging a 6B-token knowledge base, it prompts the LLM to generate professional training guidance by retrieving domain-specific QA pairs. Experimental results demonstrate that MotionDTW significantly outperforms traditional methods with lower temporal error and higher IoU scores. Furthermore, ablation studies validate the KISMAM and SportsRAG, confirming that SportsGPT surpasses general LLMs in diagnostic accuracy and professionalism.
George C. Lowe, George C. Lowe, George C. Lowe
et al.
IntroductionPlayer profiling is fundamental to effective talent identification and development strategies. However, whilst anthropometric and physiological profiling is customary practice, effective evaluation of technical and tactical skills in team sports has arguable been overlooked, largely due to a lack of suitable measurement tools. Therefore, the aim of the present study was to design, validate, and test the reliability of a novel observational instrument for assessing technical and tactical skills in rugby union.MethodsThe Bangor Rugby Assessment Tool (BRAT) was developed via the following three stages: (1) completion of a targeted literature search and expert focus group to inform initial item content; (2) Bayesian structural equation modelling (BSEM) to examine instrument factor structure; and (3) establishment of instrument reliability using intraclass correlation coefficients (ICC).ResultsResults demonstrate excellent model fit (PPP = 0.511) and strong validity for both the technical and tactical factors. ICC values ranged from moderate to excellent, demonstrating good reliability (0.79).DiscussionThe assessment tool offers a valid and reliable measure of technical and tactical aptitude within rugby union, whilst maintaining the requisite practical utility valued by practitioners.
Wei-Cheng Chao, Asaduzzaman Khan, Jui-Chi Shih
et al.
Background: Chinese-Australian adolescents face unique academic and cultural challenges that may impact their lifestyle and psychological well-being. Physical activity, screen time, and sleep are known to influence well-being. However, research on the adherence to the 24-Hour Movement Guidelines among Chinese-Australian adolescents remains limited and awaits further investigation. Objective: This study hypothesized a significant positive association between adherence to the 24-Hour Movement Guidelines for physical activity, screen time, and sleep, and the psychological well-being of Chinese-Australian adolescents. Methods: A self-reported questionnaire was distributed to two language schools in Brisbane, Australia, targeting high school students from grades 7 to 12 with Chinese-Australian backgrounds. This study used multiple linear regression modelling to examine the associations between meeting or not meeting recommendations. Meeting the 24-Hour Movement Guidelines was defined as ≥60 min/day of moderate to vigorous physical activity (MVPA), ≤2 h/day of recreational screen time, and 9–11 h/night of sleep. Results: Out of 251 participants (average age: 13.31 years; 58% female), only 20.3% met two or three recommendations, while 43.3% met one, and 36.2% met none. The most common compliance was meeting only the screen time guideline alone (48%), while 9.6% met either MVPA + screen time or screen time + sleep. The regression analysis showed that meeting at least MVPA (β = 1.41, 95% CI: 0.07 to 2.74) or at least sleep (β = 1.40, 95% CI: 0.19 to 2.60) was associated with better psychological well-being. Notably, meeting MVPA and sleep guidelines was significantly associated with higher well-being (β = 3.83, 95% CI: 1.06–6.60). From the results, adherence to additional 24-Hour Movement Guidelines was associated with improved psychosocial well-being. However, a small proportion of adolescents met all the guidelines. Conclusions: Greater adherence to physical activity and sleep guidelines is linked to better psychological well-being among Chinese-Australian adolescents. These results highlight the importance of promoting healthy behaviours and implementing public health strategies to enhance education on exercise and sleep, particularly at the school and family levels, to support adolescents’ psychological well-being.
Pablo Elipe-Lorenzo, Pablo Elipe-Lorenzo, Pelayo Diez-Fernández
et al.
IntroductionDespite advances in inclusive policies and social awareness, the participation of people with disabilities (PwD) in mainstream sports remains limited due to numerous barriers. This systematic review seeks to identify and critically analyse the main obstacles hindering equitable participation of PwD in conventional sports, while proposing evidence-based strategies to overcome these challenges.MethodsFollowing PRISMA guidelines, a comprehensive search was conducted on Web of Science and SCOPUS databases, covering studies published between 2000 and 2024. After applying inclusion and exclusion criteria, 17 studies were selected for analysis.ResultsThe findings highlight major barriers, including insufficient training for coaches and sports club managers, negative and discriminatory attitudes, an entrenched ableist mindset, limited access to information, and a lack of accessible facilities. These factors collectively impede the active participation of PwD in sports.DiscussionTo overcome these challenges, a coordinated approach is essential, encompassing attitude transformation, targeted training for sports personnel, the implementation of inclusive policies, economic incentives, and enhanced communication strategies. Additional recommendations include integrating universal design principles into sports facilities, establishing support networks and fostering a cultural shift in societal perceptions of disability.
Systematic Review RegistrationPROSPERO (CRD42024544589).
With the rapid development of artificial intelligence (AI) technology, its application in college physical education has gradually shifted from marginal exploration to system integration. Based on literature review, case study and empirical data analysis, this paper systematically explores the multi-dimensional application of AI technology in college physical education, including sports performance analysis, personalized training, teaching management optimization and intelligent evaluation. The results show that AI can effectively improve teaching efficiency, scientific training and objectivity of evaluation, while promoting the precision and fairness of physical education. Through surveys and data comparisons of multiple colleges and universities, this paper verifies the significant effectiveness of AI in improving students' sports performance, reducing sports injury rates, and improving classroom organization efficiency. At the same time, the article also points out the challenges of current applications, such as infrastructure shortage, insufficient digital literacy of teachers, low algorithm transparency and data ethics concerns. Based on this, this paper proposes a path to promote the deep integration of AI into physical education from four dimensions: system construction, teacher empowerment, technology development and data governance. The study believes that AI will play the role of "intelligent engine" in future college physical education, promoting the transformation of the education paradigm from experience-driven to data-driven.
Konrad Strużek, Kornelia Karamus, Rafał Rejmak
et al.
Advances in the understanding of Alzheimer’s disease (AD) pathophysiology, along with the development of neuroimaging and biomarker analysis, have enabled the detection of neurodegenerative changes even before clinical symptoms appear. This article explores the evolution of AD diagnostic criteria, with a particular focus on the pivotal role of cerebrospinal fluid biomarkers (Aβ42, t-tau, p-tau) and brain imaging techniques (MRI, PET). The A/T/N classification system and the concept of compensatory brain mechanisms are also discussed, emphasizing their relevance in early disease detection. The modern diagnostic approach, introduced by the Dubois criteria and further developed by the NIA-AA framework, allows for the identification of AD in its preclinical phase. The presence of biomarker abnormalities in asymptomatic individuals suggests a long latent period and the activation of neuroplastic compensatory processes that may delay symptom onset. The integration of biomarkers has significantly improved diagnostic accuracy, enhanced clinical trial participant selection, and enabled more precise disease monitoring. Despite these advances, effective treatments to halt or reverse disease progression remain elusive, highlighting the urgent need for further research into compensatory mechanisms, individual variability, and early therapeutic strategies.
With the advancement of artificial intelligence, 3D human pose estimation-based systems for sports training and posture correction have gained significant attention in adolescent sports. However, existing methods face challenges in handling complex movements, providing real-time feedback, and accommodating diverse postures, particularly with occlusions, rapid movements, and the resource constraints of Internet of Things (IoT) devices, making it difficult to balance accuracy and real-time performance. To address these issues, we propose GTA-Net, an intelligent system for posture correction and real-time feedback in adolescent sports, integrated within an IoT-enabled environment. This model enhances pose estimation in dynamic scenes by incorporating Graph Convolutional Networks (GCN), Temporal Convolutional Networks (TCN), and Hierarchical Attention mechanisms, achieving real-time correction through IoT devices. Experimental results show GTA-Net's superior performance on Human3.6M, HumanEva-I, and MPI-INF-3DHP datasets, with Mean Per Joint Position Error (MPJPE) values of 32.2mm, 15.0mm, and 48.0mm, respectively, significantly outperforming existing methods. The model also demonstrates strong robustness in complex scenarios, maintaining high accuracy even with occlusions and rapid movements. This system enhances real-time posture correction and offers broad applications in intelligent sports and health management.
Jersey number recognition is an important task in sports video analysis, partly due to its importance for long-term player tracking. It can be viewed as a variant of scene text recognition. However, there is a lack of published attempts to apply scene text recognition models on jersey number data. Here we introduce a novel public jersey number recognition dataset for hockey and study how scene text recognition methods can be adapted to this problem. We address issues of occlusions and assess the degree to which training on one sport (hockey) can be generalized to another (soccer). For the latter, we also consider how jersey number recognition at the single-image level can be aggregated across frames to yield tracklet-level jersey number labels. We demonstrate high performance on image- and tracklet-level tasks, achieving 91.4% accuracy for hockey images and 87.4% for soccer tracklets. Code, models, and data are available at https://github.com/mkoshkina/jersey-number-pipeline.