A Random Forest approach to identify metrics that best predict match outcome and player ranking in the esport Rocket League.
Ontology highlight
ABSTRACT:
SUBMITTER: Smithies TD
PROVIDER: S-EPMC8481284 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:CONTEXT:Postconcussion deficits in neurocognitive performance and postural control may persist at the time of return to sport participation. How these deficits, if present, affect athletic performance is largely unknown, with prior studies showing mixed results. OBJECTIVE:To evaluate postconcussion National Hockey League player performance using advanced hockey metrics over short- (5 games), medium- (10 games), and long-term (remainder of the season) seasonal performance. DESIGN:Retrospective cohort study. PATIENTS OR OTHER PARTICIPANTS:National Hockey League players who sustained a sport-related concussion (SRC; n = 93) and returned during the same season and players (n = 51) who missed time for non-injury-related reasons. MAIN OUTCOME MEASURE(S):Six performance metrics were used: (1) points per 60 minutes, (2) Corsi percentage, (3) personal Fenwick shooting percentage, (4) scoring chances per 60 minutes, (5) penalty difference, and (6) PDO (not an acronym but sometimes referred to as SVSP% [save percentage shooting percentage]). Performance was compared using 2 (group) × 2 (time) repeated-measures analyses of variance for 3 time windows: (1) ±5 games, (2) ±10 games, and (3) the remainder of the season postconcussion. Alpha values were set at a conservative .01 to account for the lack of independence among dependent variables. RESULTS:No significant interactions were present for any of the 6 dependent variables at any of the 3 time windows. Overall, none of the secondary variables differed. CONCLUSIONS:Using advanced, sport-specific metrics, we found that National Hockey League players did not display worse seasonal performance during 3 postinjury time frames after they sustained an SRC. Whereas laboratory studies have identified lingering neurologic deficits after concussion, our results suggest that these deficits, if present, either do not translate to worse athletic performance or were not captured by these 44 metrics. Further, prospective efforts are needed to accurately quantify performance after SRC among professional hockey players.
| S-EPMC6602367 | biostudies-literature
Project description:BACKGROUND:In ab initio protein-structure predictions, a large set of structural decoys are often generated, with the requirement to select best five or three candidates from the decoys. The clustered central structures with the most number of neighbors are frequently regarded as the near-native protein structures with the lowest free energy; however, limitations in clustering methods and three-dimensional structural-distance assessments make identifying exact order of the best five or three near-native candidate structures difficult. RESULTS:To address this issue, we propose a method that re-ranks the candidate structures via random forest classification using intra- and inter-cluster features from the results of the clustering. Comparative analysis indicated that our method was better able to identify the order of the candidate structures as comparing with current methods SPICKR, Calibur, and Durandal. The results confirmed that the identification of the first model were closer to the native structure in 12 of 43 cases versus four for SPICKER, and the same as the native structure in up to 27 of 43 cases versus 14 for Calibur and up to eight of 43 cases versus two for Durandal. CONCLUSIONS:In this study, we presented an improved method based on random forest classification to transform the problem of re-ranking the candidate structures by an binary classification. Our results indicate that this method is a powerful method for the problem and the effect of this method is better than other methods.
| S-EPMC6929337 | biostudies-literature
Project description:Polyculture is a potentially interesting rearing practice for future aquaculture developments. Nevertheless, it may result in beneficial as well as detrimental consequences for fish production. One way to maximize the benefits of polyculture is to combine species with high levels of compatibility and complementarity. This requires the development of a ranking procedure, based on a multi-trait assessment, that highlights the most suitable species combinations for polyculture. Moreover, in order to ensure the relevance of such a procedure, it is important to integrate the socio-economic expectations by assigning relative weights to each trait according to the stakeholder priorities. Here, we proposed a ranking procedure of candidate fish polycultures (i.e., species combinations that could be potentially interesting for aquaculture) based on a multi-trait assessment approach and the stakeholder priorities. This procedure aims at successively (i) weighting evaluation results obtained for each candidate polyculture according to stakeholder priorities; (ii) assessing differentiation between candidate species combinations based on these weighted results; and (iii) ranking differentiated candidate polycultures. We applied our procedure on three test cases of fish polycultures in recirculated aquaculture systems. These test cases each focused on a target species (two on Sander lucioperca and one on Carassius auratus), which were reared in two or three different alternative candidate fish polycultures. For each test case, our procedure aimed at ranking alternative combinations according to their benefits for production and/or welfare of the target species. These benefits were evaluated based on survival rate as well as morphology, behavioral, and physiological traits. Three scenarios of stakeholder priorities were considered for weighting evaluation results: placing a premium on production, welfare, or both for the target species. A comparison of our procedure results between these scenarios showed that the ranking changed for candidate polycultures in two test cases. This highlights the need to carefully consider stakeholder priorities when choosing fish polycultures.
| S-EPMC9502242 | biostudies-literature
Project description:Resistance prediction and mutation ranking are important tasks in the analysis of Tuberculosis sequence data. Due to standard regimens for the use of first-line antibiotics, resistance co-occurrence, in which samples are resistant to multiple drugs, is common. Analysing all drugs simultaneously should therefore enable patterns reflecting resistance co-occurrence to be exploited for resistance prediction. Here, multi-label random forest (MLRF) models are compared with single-label random forest (SLRF) for both predicting phenotypic resistance from whole genome sequences and identifying important mutations for better prediction of four first-line drugs in a dataset of 13402 Mycobacterium tuberculosis isolates. Results confirmed that MLRFs can improve performance compared to conventional clinical methods (by 18.10%) and SLRFs (by 0.91%). In addition, we identified a list of candidate mutations that are important for resistance prediction or that are related to resistance co-occurrence. Moreover, we found that retraining our analysis to a subset of top-ranked mutations was sufficient to achieve satisfactory performance. The source code can be found at http://www.robots.ox.ac.uk/~davidc/code.php.
| S-EPMC7188832 | biostudies-literature
Project description:BackgroundRugby league is a collision sport, where players are expected to be physically competent in a range of areas, including aerobic fitness, strength, speed and power. Several studies have attempted to characterise the physical demands of rugby league match-play, but these studies often have relatively small sample sizes based on one or two clubs, which makes generalisation of the findings difficult. Therefore, the aim of this review was to synthesise studies that investigated the physical demands of professional rugby league match-play.MethodsSPORTDiscus, CINAHL, MEDLINE (EBSCO) and Embase (EBSCO) databases were systematically searched from inception until October 2018. Articles were included if they (1) recruited professional rugby league athletes aged ≥ 18 years and (2) provided at least one match-play relevant variable (including playing time, total and relative distance, repeat high-intensity efforts (RHIE), efforts per RHIE, accelerations and decelerations, total and relative collisions). Meta-analyses were used to provide pooled estimates ± 95% confidence intervals.ResultsA total of 30 studies were included. Pooled estimates indicated that, compared to adjustables and backs, forwards have less playing time (- 17.2 ± 5.6 and - 25.6 ± 5.8 min, respectively), cover less 'slow-speed' (- 2230 ± 735 and - 1348 ± 655 m, respectively) and 'high-speed' distance (- 139 ± 108 and - 229 ± 101 m, respectively), but complete more relative RHIEs (+ 0.05 ± 0.05 and + 0.08 ± 0.04 per minute, respectively), and total (+ 12.0 ± 8.1 and + 12.8 ± 7.2 collisions, respectively) and relative collisions (+ 0.32 ± 0.22 and + 0.41 ± 0.22 collisions per minute, respectively). Notably, when the distance was expressed relative to playing time, forwards were not different from adjustables and backs in slow-speed (P ≥ 0.295) and high-speed (P ≥ 0.889) relative distance. The adjustables and backs subgroups were similar in most variables, except playing time (shorter for adjustables, - 8.5 ± 6.2 min), slow-speed distance (greater for adjustables, + 882 ± 763 m) and total relative distance (greater for adjustables, + 11.3 ± 5.2 m·min-1). There were no significant differences between positional groups for efforts per RHIE, accelerations and decelerations (P ≥ 0.745).ConclusionsThese results indicate the unique physical demands of each playing position and should be considered by strength and conditioning and tactical coaches when planning for professional rugby league performance. PROTOCOL REGISTRATION: https://osf.io/83tq2/.
| S-EPMC6560119 | biostudies-literature
Project description:This study aimed to develop a model to objectively benchmark professional Australian Rules football (AF) player performance based on age, experience, positional role and both draft type and round in the Australian Football League (AFL). The secondary aims were to identify the stage of peak performance and specific breakpoints in AF player performance longitudinally. AFL Player Ratings data were obtained for all players (n = 1052) from the 1034 matches played during the 2013-2017 seasons, along with data pertaining to the abovementioned player characteristics. Two separate linear mixed models revealed that all factors influenced player performance, with age and experience the strongest in each model, respectively. Post hoc Tukey tests indicated that performance was affected by age at each level up until the age of 21 (effect ranging from 0.98 to 3.70 rating points), and by experience at the levels 1-20 and 21-40 matches in comparison to all higher levels of experience (effect ranging from 1.01 to 3.77 rating points). Two segmented models indicated that a point of marginal gains exists within longitudinal performance progression between the age levels 22 and 23, and the experience levels 41-60 and 61-80 matches. Professional sporting organisations may apply the methods provided here to support decisions regarding player recruitment and development.
| S-EPMC6554430 | biostudies-literature
Project description:PubMed is a free search engine for biomedical literature accessed by millions of users from around the world each day. With the rapid growth of biomedical literature-about two articles are added every minute on average-finding and retrieving the most relevant papers for a given query is increasingly challenging. We present Best Match, a new relevance search algorithm for PubMed that leverages the intelligence of our users and cutting-edge machine-learning technology as an alternative to the traditional date sort order. The Best Match algorithm is trained with past user searches with dozens of relevance-ranking signals (factors), the most important being the past usage of an article, publication date, relevance score, and type of article. This new algorithm demonstrates state-of-the-art retrieval performance in benchmarking experiments as well as an improved user experience in real-world testing (over 20% increase in user click-through rate). Since its deployment in June 2017, we have observed a significant increase (60%) in PubMed searches with relevance sort order: it now assists millions of PubMed searches each week. In this work, we hope to increase the awareness and transparency of this new relevance sort option for PubMed users, enabling them to retrieve information more effectively.
| S-EPMC6112631 | biostudies-literature
Project description:We study the number of causal variants and associated regions identified by top SNPs in rankings given by the popular 1 df chi-squared statistic, support vector machine (SVM) and the random forest (RF) on simulated and real data. If we apply the SVM and RF to the top 2r chi-square-ranked SNPs, where r is the number of SNPs with P-values within the Bonferroni correction, we find that both improve the ranks of causal variants and associated regions and achieve higher power on simulated data. These improvements, however, as well as stability of the SVM and RF rankings, progressively decrease as the cutoff increases to 5r and 10r. As applications we compare the ranks of previously replicated SNPs in real data, associated regions in type 1 diabetes, as provided by the Type 1 Diabetes Consortium, and disease risk prediction accuracies as given by top ranked SNPs by the three methods. Software and webserver are available at http://svmsnps.njit.edu.
| S-EPMC3089490 | biostudies-literature
Project description:Competition is ubiquitous in many complex biological, social, and technological systems, playing an integral role in the evolutionary dynamics of the systems. It is often useful to determine the dominance hierarchy or the rankings of the components of the system that compete for survival and success based on the outcomes of the competitions between them. Here we propose a ranking method based on the random walk on the network representing the competitors as nodes and competitions as directed edges with asymmetric weights. We use the edge weights and node degrees to define the gradient on each edge that guides the random walker towards the weaker (or the stronger) node, which enables us to interpret the steady-state occupancy as the measure of the node's weakness (or strength) that is free of unwarranted degree-induced bias. We apply our method to two real-world competition networks and explore the issues of ranking stabilization and prediction accuracy, finding that our method outperforms other methods including the baseline win-loss differential method in sparse networks.
| S-EPMC4269436 | biostudies-literature
Project description:Increasing numbers of young video gamers view esports (i.e., competitive video gaming) as a career opportunity, rather than just a recreational activity. Previous studies have explored the motivational differences between esport and recreational gamers and the motivational changes through career journey to become a professional esport player. The present study explored the predictors of career plans to become a professional esport player, with a specific focus on gaming motivations. Gaming time, gaming motivations, and esport-related playing experience were also examined among Hungarian gamers with competitive gaming experience (N = 190), such as years spent in esports, medium and frequency of participating in esport tournaments, the effort put into training before the tournaments, and the plans to become a professional esport player. Binary logistic regressions were carried out and results showed that the gaming motivations of competition, skill development, and social motivations predicted career planning as a professional esport player. Additionally, results showed that younger players were more likely to seek career opportunity as professional esport players than older players. Future studies should focus on novice esport players' psychological exposure to the hypercompetitive scene of esports, such as high expectations or the risk of becoming problematic videogame users due to their motivational changes.
| S-EPMC7438909 | biostudies-literature