Machine and deep learning for sport-specific movement recognition: a systematic review of model development and performance

Objective assessment of an athlete’s performance is of importance in elite sports to facilitate detailed analysis. The implementation of automated detection and recognition of sport-specific movements overcomes the limitations associated with manual performance analysis methods. The object of this study was to systematically review the literature on machine and deep learning for sport-specific movement recognition using inertial measurement unit (IMU) and, or computer vision data inputs. A search of multiple databases was undertaken. Included studies must have investigated a sport-specific movement and analysed via machine or deep learning methods for model development. A total of 52 studies met the inclusion and exclusion criteria. Data pre-processing, processing, model development and evaluation methods varied across the studies. Model development for movement recognition were predominantly undertaken using supervised classification approaches. A kernel form of the Support Vector Machine algorithm was used in 53% of IMU and 50% of vision-based studies. Twelve studies used a deep learning method as a form of Convolutional Neural Network algorithm and one study also adopted a Long Short Term Memory architecture in their model. The adaptation of experimental set-up, data pre-processing, and model development methods are best considered in relation to the characteristics of the targeted sports movement(s).     

On Collection Size and Retrieval Effectiveness

The relationship between collection size and retrieval effectiveness is particularly important in the context of Web search. We investigate it first analytically and then experimentally, using samples and subsets of test collections. Different retrieval systems vary in how the score assigned to an individual document in a sample collection relates to the score it receives in the full collection; we identify four cases.We apply signal detection (SD) theory to retrieval from samples, taking into account the four cases and using a variety of shapes for relevant and irrelevant distributions. We note that the SD model subsumes several earlier hypotheses about the causes of the decreased precision in samples. We also discuss other models which contribute to an understanding of the phenomenon, particularly relating to the effects of discreteness. Different models provide complementary insights.Extensive use is made of test data, some from official submissions to the TREC-6 VLC track and some new, to illustrate the effects and test hypotheses. We empirically confirm predictions, based on SD theory, that P@n should decline when moving to a sample collection and that average precision and R-precision should remain constant. SD theory suggests the use of recall-fallout plots as operating characteristic (OC) curves. We plot OC curves of this type for a real retrieval system and query set and show that curves for sample collections are similar but not identical to the curve for the full collection.     

The Methodology of the 'Views and Visions Project' of Land Use-UK: Practices,issues, decisions and experiences

The Land Use-UK Project was conducted with the support of teachers and students in schools throughout the UK. At the time they acted as surveyors of the landscape and, as well as mapping the land use, students recorded their perceptions of what they had seen together with their views for the future. Referred to now as the 'Views and Visions Project', these collective opinions from mostly whole class groups of students provide a database of 1287 responses. Along with the richness of discovery and meaning which the readings of these narratives of landscape experience provide, there have been many decisions to be made regarding the appropriate style and strategies needed for development of the data and interpretation and analysis of the research. This paper provides a reflection on the methodological decisions taken from the personal perspective of the authors as well as some critical comment on the limitations and possibilities of this essentially qualitative research approach.     

Preserving Content from Your Institutional Repository

Between institutional repositories and hosting journals, many libraries are becoming responsible for scholarly content in new ways. While Portable Document Formats (PDFs) are the most common formats today, the unique, local, serial content could be in a variety of formats. These items might be digitized text, born digital text, audio, video, images, or multimedia. This article discusses formats that will remain accessible through time (PDF/A, txt, xml) so that contents are not locked into proprietary formats. It will also discuss options for backing up items and associated metadata, including simple backups, off-site storage of files, Lots of Copies Keep Stuff Safe (LOCKSS), Private LOCKSS Networks, and Portico. The article offers suggestions for how your library might best preserve local content.     

Match running performance and skill execution improves with age but not the number of disposals in young Australian footballers

This study investigated age-related differences in maturity, physical and functional characteristics and playing performance in youth Australian Football (AF). Young male players (n = 156) were recruited from 12 teams across 6 age groups (U10–U15) of a recreational AF club. All players were tested for body size, maturity and fitness. Player performance was assessed during a match in which disposals (kicks and handballs) and their effectiveness were coded from a video recording and match running performance measured using Global Positioning System. Significant main effects (P < 0.01) for age group were observed for age, years to peak height velocity, body mass, height, 20 m sprint, maximal speed over 20 m, vertical jump, 20 m multistage shuttle run, match distance, high-speed running distance, peak speed, number of effective disposals and percentage of effective disposals. Age-related differences in fitness characteristics (speed, lower body power and endurance) appeared to transfer to match running performance. The frequency in which players disposed of the football did not differ between age groups, however the effectiveness of each disposal (i.e., % effective disposals) improved with age. Match statistics, particularly those that evaluate skill execution outcome (i.e., effectiveness), are useful to assess performance and to track player development over time. Differences between age groups, and probably variability within age groups, are strongly associated with chronological age and maturity.     

Reading: A Pastime of the Past?

It is somewhat of a truism to say that reading is not what it used to be. ‘Texts’for reading now include many other media than simply print on pages and this fact has occasioned some concern among those who fear that reading television or computer-based texts may reduce or impoverish children's overall reading experience. Chris Robertson and her students have been investigating these issues through interviews with children and have found that reading and technology are not actually in conflict but can be mutually supportive. This article suggests that teachers need to think more creatively about how to use alternative texts in developing children's reading.