iSports: A web-oriented expert system for talent identification in soccer

Louzada, Francisco; Maiorano, Alexandre Cristovão; Ara, Anderson

doi:10.1016/j.eswa.2015.09.007

Cited by 31 publications

(20 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…20 The present model was successfully applied in this sample, and is currently being processed to be widely applied, by the assessment of all the necessary factors and variables to cover the most relevant aspects, even now, a software for talent selection is being generated, which will be made available to sports coaches and teachers. 21 The proposed model is far from being contemplated with all the necessary factors and variables covering all the necessary aspects. This study fulfilled three initial steps to generate a "Biological Model of Detection, Prescription and Prognosis" of future sports talents, which according to the Brazilian model are: a.…”

Section: Discussionmentioning

confidence: 99%

Detecting Sporting Talents With Z-Strategy - Cross Sectional Study

Souza-Lima

Leppe

Yáñez-Sepúlveda

et al. 2020

Rev Bras Med Esporte

View full text Add to dashboard Cite

Introduction: Due to the relationship between early identification of physical and anthropometric characteristics above the population mean in children and adolescents, and success in sports, detecting potential sports talents should be broadly and systematically used as a strategy for the early identification of physical characteristics favorable to the sport in question. However, most studies do not use representative samples, or else they present talent detection without using valid scientific methods. This retrospective, comparative study therefore presents the identification of potential sports talents using the Z Strategy, calculated with anthropometric, neuromotor and physical fitness data. Objective: To identify physical abilities and anthropometric values above what are considered the normal ranges in a population of students in the 8th year of basic education, in Chile. Methods: The sample consisted of 9,429 students from public and private schools (50.9% boys). Data were obtained from a cross-sectional study conducted in 2013. Physical fitness and anthropometric data were recompiled through the Educational Quality Measurement System (Sistema de Medición de la Calidad de Educación -SIMCE) of physical education. The "Z Strategy" was used to detect sports talents by identifying values above the population mean. Results: In at least one variable, a total of 619 male and 623 female students with a standard deviation ≥2 (Z2) were detected. Conclusion: "Z Strategy" was able to detect sports talents of both sexes and of different ages. Level of evidence III; Retrospective comparative study. Z" fue utilizada para detectar los talentos deportivos al identificar los valores por encima del promedio de la población. Resultados: Se detectó en por lo menos una variable, un total de 619 alumnos y 623 alumnas con una desviación estándar de ≥2 (Z2). Conclusión: La "Estrategia-Z" fue capaz de detectar talentos deportivos de ambos sexos y en diferentes edades. Nivel de evidencia III; Estudio retrospectivo comparativo.

show abstract

Section: Discussionmentioning

confidence: 99%

Detecting Sporting Talents With Z-Strategy - Cross Sectional Study

Souza-Lima

Leppe

Yáñez-Sepúlveda

et al. 2020

Rev Bras Med Esporte

View full text Add to dashboard Cite

show abstract

“…Nieuwenhuis et al (2002) reported accuracy of 90.5% when young female field hockey players were classified as successful or less successful. A web-oriented expert system for talent identification in soccer was developed by Louzada, Maiorano, and Ara (2016). It applies principal component and factor analysis to compute general scores for the players in real time.…”

Section: Discussionmentioning

confidence: 99%

Talent identification in soccer using a one-class support vector machine

Jauhiainen

Äyrämö

Forsman³

et al. 2019

International Journal of Computer Science in Sport

View full text Add to dashboard Cite

Identifying potential future elite athletes is important in many sporting events. The successful identification of potential future elite athletes at an early age would help to provide high-quality coaching and training environments in which to optimize their development. However, a large variety of different skills and qualities are needed to succeed in elite sports, making talent identification generally a complex and multifaceted problem. Due to the rarity of elite athletes, datasets are inherently imbalanced, making classical statistical inference difficult. Therefore, we approach talent identification as an anomaly detection problem. We trained a nonlinear one-class support vector machine (one-class SVM) on a dataset (N=951) collected from 14-year-old junior soccer players to detect potential future elite players. The mean area under the receiver operating characteristic curve (AUC-ROC) over the tested hyperparameter combinations was 0.763 (std 0.007). The most accurate model was obtained when physical tests, measuring, for example, technical skills, speed, and agility, were used. According to our results, the proposed approach could be useful to support decision-makers in the process of talent identification.

show abstract

“…Most of the TI had measured and selected young athletes based on the performance of genetically driven by a specific component [3]. It is because talent development and identification is a multifactorial process involving many characteristics.…”

Section: Introductionmentioning

confidence: 99%

Determining Youth Profile using Principle Component Analysis for Identifying Talent in Sports

2019

ijrte

View full text Add to dashboard Cite

This study attempts to determine youth profile in sports talent identification program. Data of anthropometric and physical fitness included power, agility, speeds, flexibility, strength and endurance were obtained from 600 participants in a sports talent identification program aged 13-15 years. Data analyses were carried out using multivariate analysis of principal component analysis (PCA). PCA revealed three principal components with 71.5% total variation for this studied group. The first component consists of high factor loading in speed variables (10 meter run, 20 meter run, 40 meter run) and endurance (predicted VO₂ max). The second component was constituted by anthropometric variables (weight, height, sitting height and armspan). While third component contains flexibility variable (sit and reach). These selected variables of anthropometric and fitness are, therefore, revealed as the essential attributes those must be prioritized for a talent scouting in sports. Present results had demonstrated youth profile that capable of providing an information that could help coaches in decision making during athlete selection in sports talent identification.

show abstract

iSports: A web-oriented expert system for talent identification in soccer

Cited by 31 publications

References 37 publications

Detecting Sporting Talents With Z-Strategy - Cross Sectional Study

Detecting Sporting Talents With Z-Strategy - Cross Sectional Study

Talent identification in soccer using a one-class support vector machine

Determining Youth Profile using Principle Component Analysis for Identifying Talent in Sports

Contact Info

Product

Resources

About