Dissertations / Theses on the topic 'Soccor'

Recently, injury prevention exercise programs (IPEPs) for soccer have received considerable attention and their efficacy has been demonstrated in large-scale trials. However, the ultimate impact of IPEPs will depend not only on their efficacy under controlled conditions, but also on the extent to which they are successfully implemented under real-world conditions. Despite increasing recognition of the challenges involved in successfully implementing IPEPs, there is a paucity of research addressing these challenges. The first aim of this thesis was to systematically review published IPEP trial reports, from an implementation perspective, in both soccer and other team ball sports. To achieve this, an established health-promotion framework, called the Reach Effectiveness Adoption Implementation Maintenance (RE-AIM) framework, was employed. The subsequent phases of the thesis aimed to identify IPEP implementation barriers and facilitators in the specific context of professional soccer. Two cross-sectional surveys explored the perceptions of IPEP users in professional teams and a prospective observational study assessed IPEP use, over one playing season, in a professional youth soccer academy. Evaluation of the published literature against the RE-AIM framework revealed major gaps in the reporting of specific IPEP implementation aspects, particularly relating to program adoption and maintenance. In professional soccer teams, multiple IPEP implementation barriers and facilitators were identified. These factors related either to the content and nature of the IPEPs themselves (e.g. variation, progression and soccer-specificity), or the delivery and support of programs (e.g. communication and team work) at different levels of the professional soccer ecology. In summary, there are major gaps in the reporting of implementation aspects in team ball sport trials. To enhance the implementation of IPEPs in professional soccer settings, the content and delivery of programs require significant tailoring to the specific implementation context.
Doctor of Philosophy

The aim of this thesis was to explore the role of soccer in the personal development of socio-economically disadvantaged individuals in the UK. Despite the adage that soccer is an effective vehicle for positive youth development and for diverting young people from delinquency and crime, little empirical evidence exists concerning how and why soccer can help transform the lives of disadvantaged individuals. Accordingly, this thesis employed a range of qualitative and longitudinal quantitative designs to address these gaps in the existing literature. Study one employed life story methods to explore the role of soccer in the life of coach who works with disadvantaged youth. The subsequent studies adopted the theoretical framework of self-determination theory (SOT; Deci & Ryan, 2000), to examine the coaching and psychological processes within an organised soccer programme for disadvantaged individuals. This thesis revealed the significance of soccer in the lives of disadvantaged individuals. Collectively, the results highlight the importance of organised sport and education programmes in the personal development of socio-economically disadvantaged individuals. More specifically, the role of the coach within these programmes was found to be particularly important in this development. The thesis presents key practical implications for the development and delivery of sport and education programmes for disadvantaged individuals and makes several advancements of SOT.

There have been an increasing interest in combining artifical intelligence with sports in the last decade. In the Robocup Simulation League soccer is simulated and a championship is held annually where several multi agent systems compete. What similarities and differences can be found between simulated soccer and real soccer? How is an effective team created? This essay examines methods of improvement by analyzing previous research and suggest new possible solutions. The results of these litterary studies, identified similarities in the importance of role distributions on the field and the need for efficient tactics. The most significant difference found is that localization is a task that requires more effort in a simulation environment. Several methods implemented in successful teams were identified and findings from this essay made several contributions to the current litterature.
Intresset att kombinera artificiell intelligens med sport har ökat under det senaste decenniet. I Robocup Simulation League simuleras fotboll och ett mästerskap hålls årligen där flera multi-agent system tävlar. Vilka likheter och skillnader finns mellan simulerad fotboll och verklig fotboll? Hur skapas ett effektivt lag? Denna uppsats undersöker metoder för förbättring genom att analysera tidigare forskning och föreslå nya möjliga lösningar. Resultaten av dessa litterär studier identifierade likheter i vikten av rolldistributioner på fältet och behovet av en effektiv taktik. Den viktigaste skillnaden som identifierades är att lokalisering är en uppgift som kräver mer ansträngning i en simuleringsmiljö. Flera metoder som implementerats i framgångsrika team identifierades och resultaten från denna uppsats gjorde flera bidrag till den aktuella litteraturen.

The game of soccer places various physiological demands on players, who are required to respond by carrying out a range of locomotor activities at different intensities. Such activity patterns contribute to a high energy turnover in both training and match-play, which in turn requires the intake of adequate fuel sources to sustain it. Adequate nutrient intake constitutes an important foundation for physical performance by providing fuel for biological work, both short and long term as required throughout the course of a soccer match. Despite the popularity of the sport worldwide, few studies investigating the effect of nutrition on performance in soccer have been conducted to substantiate the effect of optimal nutrition over the duration of a standard soccer match. This dearth of evidence has contributed in turn to a lack of specific nutrition guidelines for soccer players, coaches, sport scientists, nutritionists and club administrators. The aim of this thesis is to investigate the physiological, psychological and sociological aspects of nutrition and soccer, and posit a comprehensive nutritional framework to actively engage soccer players in adopting diets supportive of optimal performance in soccer training and match-play. The researcher set out by hypothesising that an optimal diet positively influences physical performance in soccer match-play. To test this hypothesis, a 90 h diet was developed for the participating semi-professional soccer players. Following implementation of the diet, a 90-minute soccer-specific simulation protocol, BEAST90mod was used to test its effects on physical performance in soccer. The remaining three inter-linked studies were longitudinal in nature, and carried out with participants forming part of the Malta U21 National Soccer Team. In the first instance, players’ habitual dietary intake, expenditure and energy balance was examined. The researcher then gathered information about the players’ knowledge, attitudes, habits, perceptions and barriers towards a diet conducive with optimal soccer performance. The same players finally underwent a 9-month nutritional education and support intervention, the efficacy of which was measured by the researcher throughout the intervention period. The principle aim of the research is to present findings that provide players and stakeholders in soccer a clear indication of the effects of specialist nutrition in soccer, and empower them with a range of appropriate tools and strategies as employed throughout the support programme. It ultimately seeks to improve physical performance in soccer training and match-play by informing sound individual and team approaches to nutritional decision-making.

Thesis (MSc (Physiotherapy))--University of Limpopo (Medunsa Campus), 2009
Aim: The aim of the study was to review the current management of soccer injuries sustained by professional players and how injury recurrence is prevented among Premier Soccer League (PSL) players in RSA. Study design: Cross sectional study, descriptive survey was followed, as it allowed the researcher to observe and describe the factors that had contributed towards recurrent injuries, as well as a qualitative approach with open ended questions where participant’s expressions encountered during the study, were analyzed. Population: The population of the study included all the medical team members in the PSL teams. PSL has 16 teams. The medical team comprised of any combination of the following members: the team doctor, physiotherapist, masseur, biokineticists, physical trainer, chiropractors and podiatrist. Not all teams have the same combination nor all disciplines mentioned. Sample: Purposive sampling method was used, as the researcher selected main participants with a specific purpose. Each of the PSL teams has 7 members from different disciplines therefore, all available members of the medical team of each PSL teams participated in the study. Data Analysis: Data was analyzed using SPPSS version 13.0.Data collected was loaded onto the statistica program for processing. Frequency tables for all variables are given to demonstrate the distribution of variables. Cross tabulation was also done to establish the relationship between the predictor variables and the response. The Test Statistic CHI-Squared (χ 2 ) was calculated and pvalue of less than 0.05 was determined for the significance. Frequency distribution tables, graphs, and charts were used to analyze the data collected. Content analysis was used for qualitative data. Results: Main findings of the study reflected that the ankle was the most commonly affected body part (69.2%) while the quadriceps and tendon achillies were both less commonly affected v (15.4%). The groin was found to incur most recurring injuries (69.2%). The results revealed that injured players are adequately managed by the medical team (61.5%). Attainment of pain free and full range of movement is the consideration for return to sport however less consideration was given to the persistent biomechanical abnormality. Conclusion: The study confirms that soccer injuries amongst Premier Soccer League (PSL) teams are managed by qualified professionals. The attainment of a free, full range of movement and adequate strength are used as criteria for return to sport and as a measure of preventing injury recurrence. Factors such as players missing out on rich rewards, recognition and regaining position in the team seem to have a great influence as well. Recommendation: The primary recommendation for the player who has had multiple recurrences is to address any underlying reversible risk factors such as poor strength and poor flexibility. A session needs to be held with soccer administrators to advocate the avoidance of factors contributing to recurrent injuries.

The RoboCup 2D Soccer Simulator is an international project for the development of multiagent systems and articial intelligence in which programmed teams compete in simulated soccer. This report aims to answer the question \What is a basic team?". In order to do this, it rst identies and analyses the key components and functionalities of basic teams. Further, it presents the work of creating a fully functional team with these fundamentals implemented. In order to evaluate the created team various test matches were set up, with their results analysed. Finally the report discusses possible future work. The results from the evaluation shows that the created team is of a functional but basic standard, which suggests that the team fundamentals were well dened and that the team is a solid base for further development.
RoboCup 2D Soccer Simulator ar ett internationellt projekt for utvecklingen av multiagenta system och articiell intelligens i vilket programmerade lag tavlar i simulerad fotboll. Den har rapporten syftar till att svara pa fragan \Vad ar ett grundlaggande lag?". For att gora detta identierar den forst nyckelkomponenterna och funktionaliteterna for grundlaggande lag. Vidare redogor den arbetet i att skapa ett fullt fungerande lag med dessa fundamentala byggstenar implementerade. For att evaluera det skapade laget anordnades aven diverse testmatcher, vars resultat analyserades. Slutligen diskuterar rapporten mojligt framtida arbete. Resultatet fran evalueringen visar att det skapade laget haller en funktionell men grundlaggande standard, vilket tyder pa att de fundametala byggstenarna var valdenierade och att laget ar en solid grund for fortsatt utveckling.

CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO
Neste trabalho, apresentamos uma metodologia para obter probabilidades de classificação e rebaixamento de equipes em campeonatos de futebol. A metodologia consiste basicamente em quatro etapas. Na primeira etapa, ajustamos modelos de séries temporais para dados de contagem a séries de gols a favor e sofridos pelas equipes em partidas sucessivas do campeonato, utilizando variáveis explicativas para considerar o efeito do mando de campo, da participação de determinados jogadores e de mudanças de técnico. Alguns problemas referentes à construção de intervalos de confiança e testes de hipóteses para os hiperparâmetros dos modelos foram solucionados via bootstrap. Na segunda etapa, obtivemos as distribuições de probabilidade associadas aos resultados das partidas futuras do campeonato, utilizando o Princípio da Máxima Entropia para combinar as distribuições preditivas dos modelos ajustados. Na terceira etapa, utilizamos as distribuições dos resultados das partidas futuras para simular cenários para o campeonato e, na quarta e última etapa, estimamos as probabilidades de classificação e rebaixamento das equipes, pela freqüência relativa da ocorrência destes eventos em um grande número de cenários gerados. A metodologia foi aplicada no Campeonato Brasileiro/1999 e na Copa João Havelange/2000.
In this thesis, we develop a methodology to obtain the probabilities of qualifying and relegating of teams, in soccer championships. The methodology consists of four steps. In the first step, we fit time series models to the series of number of goals scored in soccer matches. We account for the effects of playing at home, soccer players and changes of coaches, by introducing explanatory variables. Confidence intervals and hipothesis tests are obtained by bootstrap. In the second step, we get probability distributions of the future matches results, by combining preditive distributions of the fitted models via the Maximum Entropy Principle. In the third step, we use the distributions of the matches results to generate simulation sceneries for the champhionship. In the forth and last step, we finally estimate the probabilities of qualifying and relegating of the teams, through the relative frequencies of these events, in a great number of sceneries generated. The empirical work was carried out using data from Brazilian Champhionship/1999 and João Havelange Cup/2000.

Futebol é um esporte fascinante que capta a atenção de milhões de pessoas no mundo. Equipes de futebol profissionais, bem como os meios de comunicação, têm um profundo interesse na análise de partidas de futebol. Análise estatística é a abordagem mais usada para descrever um jogo de futebol, no entanto, muitas vezes eles não conseguem captar a evolução do jogo e as mudanças de estratégias que aconteceram. Neste trabalho, apresentamos Visual Soccer Match Analysis (VSMA), uma ferramenta para a compreensão dos diferentes aspectos relacionados com a evolução de um jogo de futebol. A nossa ferramenta recebe como entrada as coordenadas de cada jogador durante o jogo, bem como os eventos associados. Apresentamos um design visual que permite identificar rapidamente padrões relevantes em jogo. A abordagem foi desenvolvida em conjunto com colegas da área da educação física com experiência em análise de futebol. Validamos a utilidade da nossa abordagem utilizando dados de várias partidas, juntamente com avaliações de especialistas.
Soccer is a fascinating sport that captures the attention of millions of people in the world. Professional soccer teams, as well as the broadcasting media, have a deep interest in the analysis of soccer matches. Statistical summaries are the most-used approach to describe a soccer match. However, they often fail to capture the evolution of the game and changes of strategies that happen. In this work, we present the Visual Soccer Match Analysis (VSMA) system, a tool for understanding the different aspects associated with the evolution of a soccer match. Our tool receives as input the coordinates of each player throughout the match and related events. We present a visual design that allows to quickly identify relevant patterns in the match. Our approach was developed in conjunction with colleagues from the physical education field with expertise in soccer analysis. We validated the system utility using several matches together with expert evaluations.

In the latter half of the twentieth century and especially in the last twenty-five years, soccer has grown exponentially in the United States. Historically, the country has been lagging behind most of the world when it comes to adoption and diffusion of the sport; however, recent studies suggest that it has been a space of exceptionalism when incorporating the participation of women.

Studies on soccer from a geographic perspective are relatively isolated and demonstrate a tendency to favor male professional athletes. There is no similar research to examine the origins of female professional soccer players. This study will contribute to filling this identified gender gap in geographic sports studies. These previous studies on male professional athletes suggest that they can geographically originate from areas of lower socioeconomic standing. The findings from this study show a distinct contrast between male and female professional athlete origins.

Results reveal that the origins of most female professional players can be connected to suburbanized middle to upper middle class white communities close to major cities mainly in coastal regions. From a per capita perspective, the results also show that states in the West produce more players than states in the East. Socio-cultural perspectives explain these patterns, supporting a common hypothesis that most female professional soccer players in the USA are white and come from areas of relative affluence.

Background: Football (Soccer) is arguably the most popular sport in the international sporting arena. A survey conducted by FIFA (Fédération International de Football Association) (FCPA, 2000) indicated that there are 240 million people who regularly play soccer around the world. Internationally, there are 300 000 clubs with approximately 1.5 million teams. In South Africa, there were 1.8 million registered soccer players in 2002/2003 (Alegi, 2004). Although youth players are predominantly amateurs and have no financial value for their clubs or schools, their continued health and safety are still of vital importance. There are some clubs which contract development players at 19 years of age in preparation for playing in their senior sides and these young players should be well looked after, to ensure a long career playing soccer. Being able to predict injuries and prevent them would be of great value to the soccer playing community. Aims: The main aim of this research was to create a statistical predictive equation combining biomechanics, balance and proprioception, plyometric strength ratios of ND/Bil (Non dominant leg plyometrics/ Bilateral plyometrics), D/Bil (Dominant leg plyometrics/ Bilateral plyometrics) and ND+D/Bil (Non dominant leg + dominant leg plyometrics/ Bilateral plyometrics) and previous injuries to determine a youth soccer player's risk of the occurrence of lower extremity injuries. In the process of reaching this aim it was necessary to record an epidemiological profile of youth soccer injuries over a two season period. It was also necessary to record a physical profile of, and draw comparisons between, school and club youth soccer players. Following the creation of the prediction model a preventative training programme was created for youth soccer players, addressing physical shortcomings identified with the model. Design: A prospective cohort study Subjects: Schoolboy players from two schools in the North West Province, as well as club players from three age groups were used for this study. Players from the U/16 and U/18 teams in the two schools were tested prior to the 2007 season. Players from the U/17, U/18 and U/19 club development teams were tested prior to the 2008 season. The combined total number of players in the teams amounted to 110 players. Method: The test battery consisted of a biomechanical evaluation, proprioceptive and plyometric testing and an injury history questionnaire. The Biomechanical evaluation was done according to the protocol compiled by Hattingh (2003). This evaluation was divided into five regions with a dysfunction score being given for each region. A single limb stance test was used to test proprioception. A Sergeant jump test was utilised using the wall mark method to test plyometric jumping height. A previous injury questionnaire was also completed on all players prior to testing. Test subjects from the schools were tested with the test battery prior to commencement of the 2007 season. The testing on the club teams was undertaken prior to the 2008 season. Injuries were recorded on the prescribed injury recording form by qualified Physiotherapists at weekly sports injury clinics at each of the involved schools and clubs. The coaching staff monitored exposure to training activities and match play on the prescribed recording forms. These training and match exposure hours were used, along with the recorded injuries for creating an epidemiological profile. Injuries were expressed as the amount of injuries per 1000 play hours. Logistical regression was done by using the test battery variables as independent variables and the variable injured/not injured as dependent variable (Statsoft, 2003). This analysis created prediction functions, determining which variables predict group membership of injured and non injured players. Results: There were 110 youth players involved in the research study from seven teams and four different age groups. There were two groups of U/16 players, an U/17 group, three U/18 groups and an U/19 group. The players were involved in a total of 7974 hours of exposure to training and match play during the seasons they were monitored. The average age of the players was 16.6 years. The majority of players were right limb dominant (83.6%) and 65.7% of players failed a single limb stance test. The mean jump height for both legs combined was 33.77cm, with mean heights of 22.60cm for dominant leg jump and 22.66cm for the non dominant leg. In the biomechanical evaluation of the lower leg and foot area, the average youth player presented with adaptation of toes, normal or flat medial foot arches, a normal or pronated rear foot in standing and lying and a normal or hypomobile mid-foot joint. Between 42.7% and 51.8% of players also presenting with decreased Achilles tendon suppleness and callusing of the transverse foot arch. The youth profile for the knee area indicated that the players presented with excessive tightness of the quadriceps muscles, normal patella tilt and squint, normal knee height, a normal Q-angle, a normal VMO: VL ratio and no previous injuries. This profile indicated very little dysfunction amongst youth players for the knee area. For the hip area, the youth profile was described as follows: There was shortening of hip external rotators, decreased Gluteal muscles length, normal hip internal rotation and no previous history of injury. Between 38.2% and 62.7% of players also exhibit shortened muscle length of the adductor and Iliopsoas muscles and decreased length of the ITB (Iliotibial Band). In the Lumbo-pelvic area there was an excessive anterior tilt of the pelvis with normal lumbar extension, side flexion, rotation and lumbar saggital view without presence of scoliosis. Between 58.18% and 65.45% of players presented with an abnormal coronal view and decreased lumbar flexion. Between 41.81% and 44.54% of players also presented with leg length, ASIS, PSIS, Cleft, Rami and sacral rhythm asymmetry. The similarity of the results for these tests in all players contributed to a new variable called 'SIJ dysfunction'. This was compiled from the average of the scores for Leg length, ASIS, PSIS, Cleft, Rami and Sacral rhythm, which was also considered for inclusion in the prediction model. The neurodynamic results of youth players indicated that approximately between 44.54% and 50.91% of players presented with decreased Straight leg raise and prone knee bend tests. The total combined dysfunction scores for the left and right sides were 17.091 and 17.909 respectively, indicating that there were higher levels of dysfunction on the right side than the left. This increased unilateral dysfunction could probably be attributed to limb dominance and increased use of the one leg for kicking and passing during the game. In the epidemiological study on youth players, there were a total of 49 training injuries and 52 match injuries. The total injury rate for youth players was 12.27 injuries/1000 hours, with a total match injury rate of 37.12 injuries/1000 match hours. The combined training injury rate was 7.17 injuries/1000 training hours. 87.13% of injuries were of the lower limb area and the individual areas with the highest percentage of injuries were the Ankle (25.74%), Knee (19.80%), Thigh (15.84%) and Lower leg (14.85%).The totals for youth players indicated that sprains (30.69% of total), strains (27.72% of total) and contusions (27.72% of total) were the most common causative mechanism of injuries. The severity of injuries show 'zero day' (no time off play) injuries to be the most common type (35.64%), followed by 'slight' (1 to 3 days off play) (33.66%) and 'minor' (4 to 7 days off play) (14.85%). School players had higher injury rates than club players but the severity of injuries to club players was higher, with longer absences from play. Non-contact injuries accounted for 52.47% of the total with 46.53% being contact injuries. School players had lower levels of non-contact injuries than club players, which correlated well with lower dysfunction scores recorded for school players during the biomechanical evaluations. This demonstrated that there was a definite relationship between levels of biomechanical dysfunction and the percentage of non-contact injuries in youth players, which formed the premise of the creation of a prediction model for non-contact youth soccer injuries. The next step in the creation of a prediction model was to identify the variables that discriminated maximally between injured and non-injured players. This was done using stepwise logistic regression analysis. After the analysis, ten variables with the largest odds ratios were selected for inclusion in the prediction model to predict non-contact injuries in youth soccer players. The prediction model created from the stepwise analysis presented as follows: P (injury)= exp(-8.2483 -1.2993a + 1.8418b + 0.2485c + 4.2850d + 1.3845e + 1.3004f-1.1566g + 1.8273h-0.9460i-0.5193j) l + exp(-8.2483-1.2993a + 1.8418b+ 0.2485c + 4.2850d + 1.3845e + 1.3004f-1.1566g + 1.8273h-0.94601-0.5193J) a = Toe dysfunction b = Previous ankle injury c = Ankle dysfunction d = SIJ dysfunction e = Lumbar Extension f = Straight Leg Raise g = Psoas length h = Patella squint i = Gluteal muscle length j = Lumbar dysfunction P = probability of non contact injury exp(x) = e x , with e the constant 2.7183 In the ankle area, the toe positional test, previous ankle injury history and combined ankle dysfunction score were included in the prediction model. In the knee area, the patella squint test was included in the model. In the hip area, the Psoas component of the Thomas test was included, along with the Gluteal muscle length test. In the Lumbo-pelvic area, the SIJ dysfunction (average of Leg length, ASIS, PSIS, Rami, Cleft and Sacral rhythm tests), lumbar extension test and lumbar dysfunction scores were included in the prediction model. In the neurodynamic area, the Straight leg raise test was included in the prediction model. The prediction model therefore contained tests from all five the bio mechanical areas of the body. Overall, this model correctly predicted 86.91% of players as either injured or not-injured. The I value (effect size index for improvement over chance) of the prediction model (1=0.67), along with the sensitivity (65.52%), specificity (94.87%), overall correct percentage of prediction (86.91%) and Hosmer and Lemeshow interferential goodness-to-fit value (X 2(8) = 0.7204), all demonstrated this prediction model to be a valid and accurate prediction tool for non-contact youth soccer injuries A second prediction model, for the prediction of hip and groin injuries amongst youth players, was also created. The prediction model created from the stepwise analysis for groin injuries presents as follows: P (Groin injury)^ exp(-116.2 + 33.5383d + 14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) l + exp(-116.2 + 33.5383d+14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) d = SIJ dysfunction k = Previous knee injury m = Previous hip injury e = Lumbar extension f = Straight leg raise n = Limb dominance p = ND/Bil plyometric ratio P = probability of groin injury exp(x) = ex, with e the constant 2.7183 The prediction model for hip and groin injuries included the variables of SIJ dysfunction, previous knee injury, previous hip injury, lumbar extension, straight leg raise, limb dominance and the ratio of non-dominant leg to bilateral legs plyometric height. When all the validifying tests were examined, the I-value (0.64868), sensitivity (66.67%), specificity (98.01%), false negatives (1.98%), false positives (33.33%), Hosmer and Lemeshow goodness-to-fit value (X2(8) = 0.77) and the overall percentage of correct prediction (96.26%) all reflected that this model was an accurate prediction tool for hip and groin injuries amongst youth soccer players. Conclusion: This study showed that it was possible to create a prediction model for non-contact youth soccer injuries based on a pre-season biomechanical, plyometric and proprioceptive evaluation along with a previous injury history questionnaire. This model appears as follows: P (injury)= exp(-8.2483 -1.2993a + 1.8418b + 0.2485c + 4.2850d + 1.3845e + 1.3004f - 1.1566g + 1.8273h - 0.9460i - 0.5193J) l + exp(-8.2483-1.2993a+ 1.8418b + 0.2485c + 4.2850d + 1.3845e + 1.3004f-1.1566g+1.8273h-0.94601-0.5193J) a = Toe dysfunction b=Previous ankle injury c = Ankle dysfunction d= SIJ dysfunction e=Lumbar Extension f = Straight Leg Raise g = Psoas length h = Patella squint i = Gluteal muscle length j = Lumbar dysfunction P = probability of non contact injury exp(x) = ex, with e the constant 2.7183 It was also possible to create a prediction model for non contact hip and groin injuries, which appears as follows: P (Groin injury)= exp(-116.2 + 33.5383d + 14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) l + exp(-116.2 + 33.5383d + 14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) d = SIJ dysfunction k = Previous knee injury m = Previous hip injury e = Lumbar extension f = Straight leg raise n = Limb dominance p = ND/Bil plyo metric ratio P = probability of groin injury exp(x) = ex, with e the constant 2.7183 It was also possible to create a prediction model for non contact hip and groin injuries, which appears as follows: P (Groin injury)= exp(-116.2 + 33.5383d + 14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) l + exp(-116.2 + 33.5383d + 14.5108k + 4.1972m + 1.9330e + 10.7006f-14.4028n + 48.8751p) d = SIJ dysfunction k = Previous knee injury m = Previous hip injury e = Lumbar extension f = Straight leg raise n = Limb dominance p = ND/Bil plyo metric ratio P = probability of groin injury exp(x) = ex, with e the constant 2.7183 Using the hip and groin prediction model, combined with the injury prediction model, injuries in youth soccer players can be predicted. The data for each player should first be substituted into the injury prediction model, to determine the chance of getting injured during the season. The data should then be substituted into the hip and groin injury prediction model, determining the chance of hip and groin injuries during the season. The results from the groin injury prediction model could then be used to exclude groin injuries amongst players. A negative result for the hip and groin injury, which showed a false negative percentage of 1.98%, could be used to determine that an injury that was predicted using the overall injury prediction model, would not be a hip and groin injury. A positive result in the groin injury test could, however, not exclude injuries to other body areas that were predicted by the overall injury prediction model, so the groin injury prediction model could only be used to exclude hip and groin injuries.
Thesis (Ph.D. (Education)--North-West University, Potchefstroom Campus, 2009.

Thesis (MTech (Tourism and Hospitality Management))--Cape Peninsula University of Technology, 2009
Football is considered one of the most important sports in several of the 53 countries in Africa, with the largest viewership and participation.

The savage essence and nature of sports means those who work on it hunt for the win. The sport enterprise is undergoing a gigantic digital transformation focused on imaging, real time and data analysis employed in the competitions. Conventional process methods in sports management such as fitness and health establishments, training, growth and match or game realisation are all being revolutionized by the sport digitization. In team sports it is well known that is needful an enough and simple digital methodology to organize and construct a feasible strategy. Digitization in sports is perpetually evolving and requires pervasive challenges. The sports and athletics digitization success is based on what is being done with collection of more data. Competitive advantages go to those who produce powerful operations using the data and acting on it in real time. The potential impact of these sport features in sport team operations is powerful. Data does not ride all decisions, but it empowers knowledgeable decisions. In these world circumstances, our vision with this system was born from a dream helping soccer sport management systems embrace and improve its contest success. Our perspective problem is how a decision support system for soccer coaches helps them to take enhancement decisions better. To face this problem we have created a soccer coach decision support system. This system is organised in two joined components; the first simulates the prediction of the soccer match winner through a data driven neural network. This component output activates the second to operate the logic rules learning and provides the stats, analysis, decision making and additionally plans improvements like drills and training procedures. This helps on the preparation towards upcoming matches as well as being aligned with their style and playing concepts. Future scalability and development, will analyse the mental and moral features of the teams by virtue of their athlete’s behavior changes.

Upper Confidence bounds applied to Trees, or UCT, has shown promise for reinforcement learning problems in different kinds of games, but most of the work has been on turn based games and single agent scenarios. In this project we test the feasibility of using UCT in an action-filled multi-agent environment, namely the RoboCup simulated soccer league. Through a series of experiments we test both low level and high level approaches. We were forced to conclude that low level approaches are infeasible, and that while high level learning is possible, cooperative multi-agent planning did not emerge.

The aim of the present research is to investigate three factors, namely perceptions of fairness in intergroup situations, ingroup identification and spatial dimensions that are assumed to contribute to why individuals participate in violence against the police in soccer stadiums. In Study 1 perceptions of fairness, identification and spatial perspective were manipulated and the results indicated a significant interaction effect between identification and spatial perspective. This interaction effect had a significant influence on negative behavioural tendencies towards police. In Study 2, identification and spatial perspective were manipulated and once again the interaction effect between identification and spatial perspective was found. A main effect of identification was found in that participants who identified lower with fans showed significantly more positive attitudes towards police. The results of the two studies highlight the importance of looking beyond the inherent nature of the crowd itself when analysing situations of police/fan conflict, and also the need to further investigate the spatial dimension and how it influences social judgment and decision making.

Soccer is the most popular ball sport in the world. With an estimated 247 million active players world-wide, the game generates an annual turnover of approximately 200 billion dollars, which is far in excess of corporate leaders such as General Motors. The soccer ball represents the fundamental equipment requirement with ball sales estimated at 40 million units per year. The market is extremely competitive and manufacturers strive for superior product performance to enable commercial advantage. The Federation Internationale de Football Association (FIFA), soccer's world governing body use quasi-static testing to scrutinise ball designs however there is a need for greater understanding of the dynamic performance characteristics of soccer balls. This thesis is directed toward the development of a modelling methodology using finite element (FE) technology supported by a set of dynamic test standards. The modelling activity has been predominantly concerned with two different ball types, the traditional manually stitched ball (MSB) which contains 32 textile reinforced composite panels pressurised by an internal rubber bladder and two new generation ball types (NGB) which incorporate an underlying 12-panel stitched fabric carcass, onto which outer panels are adhered. FE models have been created with homogeneous and isotropic material properties to simulate ball impact behaviour. Experimental measurements of coefficient of restitution, deformation, and contact time, representative of play conditions, were used to validate the models. Each ball model was developed to capture material heterogeneity, which included the valve, stitching seam for the MSB, the influential carcass structure and softer outer panel arrangement within the NGBs. Material anisotropy has been modelled in order to replicate the deformation behaviour of soccer ball s at high speeds. Anisotropy was shown to affect impact characteristics including significant variations in ball oscillations, spin, and ball motion. The validated carcass FE model was subsequently used as a predictive design tool within an optimisation process to reduce the effects of material anisotropy. This has resulted in the development of a ball with more uniform impact characteristics and a set of design guidelines for future sports ball development. Dynamic material test data has been used to define material damping for the description of kinetic energy loss throughout impact. The model combines static and dynamic force hysteresis measurements to accurately represent energy loss characteristics throughout impact. The visco-anisotropic-hyperelastic ball modelling strategy described in this thesis accurately represents the complex deformational behaviour and energy loss characteristics for soccer balls undergoing dynamic impacts. The modelling strategy has also been successfully used in a design optimisation process for the development of soccer balls with more uniform impact characteristics.

RoboCup 2D Soccer Simulation League is an international artificial intelligence (AI) competition in which computer programs compete in soccer. This report will present the work of analyzing the successful teamWrightEagle, and the isolation of the key strategic and behavioral aspects that make them successful. It will also present the work of creating a new team in which the aspects exctracted from WrightEagle has been implemented, but in a much simpler way. It was found that good passing play, good stamina preservation and the ability to stay spread out wereWrightEagle’s key strategic and behavioral aspects. While these aspects has been implemented in the created team, the lack of good core functionality proved to be a more
RoboCup 2D Soccer Simulation League är en internationell tävling i artificiell intelligens (AI), där datorprogram tävlar i fotboll. Den här rapporten kommer presentera arbetet i att analysera det framgångsrika laget WrightEagle, och att hitta de aspekter vad gäller strategi och betéende som gör dem så framgångsrika. Den kommer också att presentera skapandet av ett nytt lag som implementerar de funna aspekterna från WrightEagle, men på ett mycket enklare sätt. Bra passningsspel, bra uthållighetsbevarande, och deras förmåga att hålla sig utspridda var de funna aspekterna från WrightEagle vad gäller strategi och betéende. De här aspekterna har blivit implementerade i det skapade laget, men bristen av bra basfunktionalitet visade sig väga mycket tyngre och det resulterande laget preseterar dåligt.

The American Dream is founded upon the ideological belief that ‘you can be anything you want to be, regardless of your current class position.’ This belief is contained within the dominant prevailing notion that the U.S. is a meritocracy where power and success are associated with determination and failure with laziness. This thesis challenges whether the American Dream is a relevant, attainable and viable concept for higher education students via the avenue of a soccer scholarship. In so doing, the research presented challenges the perceived wisdom of ‘American exceptionalism’ from a critical theoretical perspective. The research question at the heart of this study is ‘what are the motives of American university students for undertaking a soccer scholarship?’ The adoption of an interpretive paradigm for this study aims to provide an explanation of student decision-making. In the final analysis, this approach reveals what soccer means to the lives of the student-athletes. The central themes of the study were established via a pilot study and categorised as: family, social class, social mobility and career development. Questionnaires were completed by 154 students from two separate Division One universities. Twelve students were then purposively sampled and interviewed using a semi-structured format. To supplement these opinions, interviews were then conducted with a selection of coaches and athletic directors at the respective institutions. Analysis of the responses was contextualised using the framework provided by Csikszentmihayli and Schneider’s (2000) ‘Support/Challenge Questionnaire’. The findings support a common hypothesis that the family is a significant agent in socialising of their children to the cultural values of the American Dream. The findings additionally reveal support for the notion that families are important influences on their child’s sport mobility orientations in the soccer context. An alternative explanation proposed here is that the transmission process is actually a two-way dialogue in which children socialised their parents and vice versa. The family in this study represent a potentially problematic social process for the inculcation of values related to the maintenance of social life. The conclusions presented clearly reveal that the majority of students embarking on a soccer scholarship are motivated by the need to firstly finance their higher education and secondly to take part in a sport they have played since childhood. Students were aware of the uncertainty of the marketplace and the limitations of their own technical ability. As such their participation in the scholarship could be considered to be a pragmatic adaptation of a ‘labour of love.’

Resisting external forces can be a source of beauty. This work represents an effort to understand how architecture can be derived from the struggle against natural forces to create a built-form. Starting with an idea, that of a tensile roof, the design of a soccer stadium developed into what is presented here.
Master of Architecture

Satisficing was introduced by the economist Herbert Simon to allow for decisions that are "good enough" when there are insufficient computational resources and knowledge to obtain the optimal outcome. Autonomous multi-agent systems often require such decision making because of the complexity and unknown factors present in such an environment. Satisficing has been extended significantly by Wynn Stirling. Through extended satisficing, he has departed from conventional approaches to autonomous multi-agent systems, based as they usually are on the assumption that each participant is motivated exclusively by its own self interest, and will therefore attempt to maximize its benefit, regardless of the benefit or cost to others. He considers an alternative view based on the assumption that, when forming its preferences, the agent is willing to take into consideration the preferences of others. This thesis explores the application of satisficing to simulated soccer, an autonomous multi-agent system with significant inherent complexity. The work described in this thesis shows that satisficing provides an easy way to switch between an agent's various roles, to take into consideration the likely goals and actions of other agents, and to work in conjunction with a genetic algorithm to help optimize parameters. Some principles of developing simple and concise satisficing code are suggested. Satisficing is thus shown to be an effective solution to decision making in complex multi-agent systems.

The purpose of this study was to discover patterns in intra-team co operation (passing) versus individualistic behavior (dribbling) across various age and gender groups . A behavioral coding scheme for soccer players in possession of the ball was designed to tabulate team responses of passing or dribbling behavior in a four-a-side indoor tournament . A total of 32 teams , 18 male teams and 14 female teams , participated in four age brackets. A multi variate analysis of variance (MANOVA) generated one significant result for age and passing behavior. This supported the first hypothesis that intra-team cooperative behavior increased with age. Due to the nonorthogonal nature of the data two separate analyses of variance (ANOVA) were conducted, one for each of_ the dependent variables. No significant results were generated by these ANOVA's for sex and dribbling behavior. However, there was tenuous confirmation of the second hypothesis, that there is a gender difference in the use of cooperative (passing) responses and individualistic (dribbling) behavior.

Philosophiae Doctor - PhD
This study presents an exploration of a group of soccer stakeholders’ perceptions and experiences of gender equity practices at four Western Cape universities in South Africa. It discusses female soccer players’ experiences of gender equity practices at universities and the implications for the structures and practices of equitable soccer organisations. The concept of soccer stakeholders in this study represents both those at leadership level, soccer administrators, and those actively participating in soccer, male and female soccer players in the universities.This study uses a feminist qualitative methodology and semi-structured individual interviews with four soccer stakeholders and 16 senior soccer players, that is, eight females and eight males, for individual interviews. In addition, focus group interviews with women only and mixed gender interviews were conducted per institution. A major finding from the study suggests the dominance of a discourse of equal and same opportunity and treatment, which was disconnected from the understandings of power, and the lack of problematising the treatment of women as ‘add-on’ to the supposedly natural and hierarchical structure of soccer. In addition, gender equity as a superficial practice and ‘favours for women’ discourses suggest the prevailing male bias in which women’s participation in soccer continues to be viewed as secondary and less valuable than men’s. Rationalising discourses for continued male dominance in soccer suggests that gender equality is ‘conditional’ for women’s teams, because the different levels of soccer they are playing at are not considered or valued the same as the men. Other key findings suggest that, firstly, the four universities did not have the gender equity policy in soccer and soccer administrators did not consider the importance of a specific and directed policy in soccer. Secondly, the results on the experiences of gender (in)equity practices in the universities corroborated the lack of support for women’s soccer and women soccer players’ experiences of marginalisation and neglect. The findings suggest that women’s soccer continues to experience inequity practices in soccer, that women’s soccer is devalued and secondary to male soccer, and that men’s soccer and men in soccer continue to be prioritised. Thirdly, the findings suggest that while the government’s sport policy is acknowledged, of concern is the lack of structures to develop soccer at grassroots level to ensure the sustainability of growth for boys’ and girls’ soccer. Furthermore, schools are identified as important institutions, particularly primary schools, to encourage and develop an interest in soccer for boys and girls, because they have paid little attention to the development of soccer for girls in different age groups, or to nurture continuity and motivation in various age groups, as compared to boys. Fourthly, there is a lack of passionate, serious, and committed people to implement and monitor the policy to make sure that the progress of gender equity practices are implemented and monitored in soccer. In addition, there is suspicion at the government’s lack of interrogating the continuing bias of the media in relation to the dominance of men’s soccer and lack of media coverage for women’s soccer. Fifth, the findings suggest that women are not playing a role in supporting their games and they need to take ownership of their soccer, because they seem to have surrendered the role of developing their soccer, and are reliant on men to develop girls’ and women’s soccer. In addition, women who play soccer reportedly experience stigmatisation through name calling and questions about their physical appearance, sexuality, and dress code. Finally, given that soccer fields are contested ‘spaces’ that have been traditionally and ‘naturally’ declared for men, research that prioritises interrogating and problematising men’s perceptions of women’s soccer and issues of gender equity practices in soccer is needed in a democratic South Africa.

Thesis (M.Sc.( Sports Physiotherapy and Rehabilitation)) --Sefako Makgatho Medical University, 2013.
Introduction: Soccer predisposes its players to a number of injuries which could adversely affect the player’s career if relevant management is not instituted. The aim of the study was to develop a profile of soccer injuries amongst the premier soccer teams in Lesotho and determine a need for physiotherapy intervention. Methodology: A cross sectional descriptive survey was used. Simple random selection was used to select a total of 201 soccer players from 12 premier teams participated in the study. Inferential statistics were performed using ANOVA, ANCOVA and Pearson’s correlation coefficient tests to determine the relationship of the players’ injuries, contributing factors and their management. Questionnaire’s validity and reliability were established by physiotherapist working with athletes. Results: The lower limb is the commonly affected body part with the ankle sustaining most injuries (36.4%). Common types of injuries were muscle strains (32.4%) and ligament sprains (23.0%). Mechanisms of injury during competitive games and training sessions include being tackled (25.2%); landing, tackling and overuse. Injuries sustained are treated sometimes by medical personnel (43.7%), traditional or home remedies (41.7%) and physiotherapist (29.6%). Ankle injuries are associated with self- treat (p= 0.020). There is a strong significant relationship between the mode of treatment (not physiotherapy) and common injuries that occurred amongst the players (p<0.05). The common mechanism of injury is also association with the common injuries that occurred (p<0.05). Conclusion and implications: Injuries occurring amongst soccer players in the Lesotho Premier Soccer League are sprains and strains and no consistent management is available for the injured players. The type and severity of common injuries indicate their impact on the soccer sporting activity. The injured player ceases activity immediately after injury. Soccer fraternity need to pay attention to provision of physiotherapy services and players have to be educated regarding appropriate injury prevention and management strategies.

Background: Three epidemiological studies conducted in Rwanda have highlighted that many people at different levels in the community of soccer do not implement accepted control measures for reducing the risk of injuries. However, little is known about what soccer community members themselves know about injury prevention. Purpose: The aim of this study was to identify perceptions of factors influencing the implementation of soccer injury prevention strategies in Rwanda.

Includes abstract.
Includes bibliographical references.
Small Size League (SSL) Soccer at RoboCup uses the complexity of a well kown human sport to extend the capabilities of mobile robotics as well as automated computer control. The game creates a environment where technically complicated miniature soccer robots are required to perform the tasks based on computer algorithms to outsmart similar systems. SSL creates a publically accessible window into complicated systems that an outsider is capable of comprehending.

An estimated 3.8 million sports-related concussions occur every year. Little research has been collected on soccer players, despite women's soccer having the third highest rate of concussion among all popular collegiate sports. The objective of this work was to evaluate multiple interventions that have been introduced to address the high rate of concussions in this population. Wearable head impact sensors were evaluated on their ability to accurately count and measure head impacts during a collegiate women's soccer season. Head impact exposure was quantified using video analysis of this season as well. Sensors were unable to accurately count impacts and reported nonsensical head acceleration measurements, indicating that data reported from head impact sensors should be interpreted with caution. The ability of soccer headgear to reduce linear and rotational head accelerations during common soccer impacts was examined in the laboratory. Ball-to-head and head-to-head impacts were performed at a range of speeds and impact orientations. Headgear resulted in small reductions during ball-to-head tests, which are not likely to be clinically relevant. In head-to-head tests, use of headgear on the struck head provided an overall 35% reduction in linear head acceleration, and a 53% reduction when another headgear was added to the striking head. The ten headgear tested varied greatly in performance. These data suggest that the use of protective headgear could reduce concussion incidence significantly in this population. Research presented in this thesis will inform soccer organizations on best practices for player safety with regard to head impacts.
Master of Science

Mestrado em Engenharia de Computadores e Telemática
O teste de algoritmos na área da robótica pode ser uma tarefa difícil, especialmente se o teste envolver múltipos robots. Neste contexto o uso de um simulador torna-se uma ferramenta importante no teste de algoritmos pois permite ultrapassar algumas limitações e oferece várias vantagens. CAMBADA é a equipa de futebol robótico da liga de tamanho médio da Universidade de Aveiro, Portugal. A equipa está familiarizada com as limitações do uso de robots reais para o teste de algoritmos. Devido a isso o simulador criado pela equipa Brainstormers Tribots foi adaptado para prover um ambiente de simulação ao software CAMBADA e estava em uso aquando do início desta dissertação. O simulador oferecia pouca flexibilidade na modelação dos robots que resultava em comportamentos imprecisos, oferecia também reduzida interacção com a simulação. O objectivo desta dissertação é criar um ambiente de simulação para agentes em futebol robótico com a intenção de melhorar o ambiente de simulação da equipa CAMBADA. O simulador deve ser capaz de simular dinâmica de objectos a três dimensões, sensores e actuadores ao mesmo tempo que oferece visualização do mundo e a possibilidade de interagir com a simulação. Da pesquisa realizada sobre simuladores robóticos o simulador Gazebo respeitava os nossos requisitos e foi escolhido para código base do nosso simulador. Para criar um ambiente simulado adequado à equipa CAMBADA alguns componentes do Gazebo foram alterados e novos sensores e actuadores virtuais foram desenvolvidos. Vários componentes do software CAMBADA tiveram que sofrer alterações de modo a suportar um ambiente simulado. O robot virtual foi modelado de modo a assemelhar-se com o robot real com o objectivo de obter comportamentos mais precisos. O simulador desenvolvido substituiu a solução anteriormente criada pela equipa CAMBADA e foi usado nos testes de preparação para a participação da equipa no RoboCup 2010 em Singapura onde deu o seu contributo na obtenção do terceiro lugar.
In the field of robotics, testing algorithms with the real robots can be a di cult task, specially if the test involves more than one robot. In this context a simulator is an important tool for testing algorithms because it helps overcome some limitation and o ers several advantages. CAMBADA is the RoboCup MSL soccer team of the University of Aveiro, Portugal. The team is familiar with the limitations of using the real robots for testing algorithms. Therefore, a simulator created by the Brainstormers Tribots team was adapted to provide a simulated environment for their software and was used for testing at the time of the beginning of this thesis. The simulator offered low flexibility on the modeling of the robots from which resulted inaccurate behaviors, it also o ered reduced interaction with the simulation. The purpose of this thesis is to create a simulation environment for robotic soccer agents with the intention of improving the simulated environment for the CAMBADA team. The simulation must provide three-dimensional dynamics of objects, be capable of simulating sensors and actuators, allow the visualization of the simulation and provide interaction with the simulation. From the conducted survey about robotic simulators, the simulator Gazebo complied with our requirements and was chosen to provide the code base for our simulator. To create an adequate simulation environment for the CAMBADA team some components of Gazebo were modi ed and new sensors and actuator were developed. Several components of the CAMBADA software had to be modified to support the simulated environment. The virtual robot was modeled to resemble the real robot to provide more accurate behaviors. The developed simulator substituted the previous solution created by CAMBADA team and was used in the preparation tests for the participation in the RoboCup 2010 in Singapore where it contributed to obtain of the third-place.

Mestrado em Engenharia de Computadores e Telemática
A soccer-playing robot must be able to carry out a set of behaviors, whose complexity can vary greatly. Manually programming a robot to accomplish those behaviors may be a difficult and time-consuming process. Automated learning techniques become interesting in this setting, because they allow the learning of behaviors based only on a very high-level description of the task to be completed, leaving the details to be figured out by the learning agent. Reinforcement Learning takes inspiration from nature and animal learning to model agents that interact with an environment, choosing actions that are more likely to lead them to accumulate rewards and avoid punishment. As agents experience the environment and the effect of their actions, they gain experience which is used to derive a policy. Agents can do this instantaneously after they observe the effect of their last action, or after collecting batches of these observations. The latter alternative, called Batch Reinforcement Learning, has been used in real world applications with very promissing results. This thesis explores the use of Batch Reinforcement Learning for learning robotic soccer behaviors, including dribbling the ball and receiving a pass. Practical experiments were undertaken with the CAMBADA simulator, as well as with the CAMBADA robots.
Um robô futebolista necessita de executar comportamentos variados, desde os mais simples aos mais complexos e difíceis. Programar manualmente a execução destes comportamentos pode tornar-se uma tarefa bastante morosa e complicada. Neste contexto, os métodos de aprendizagem automática tornam-se interessantes, pois permitem a aprendizagem de comportamentos através de uma especificação a muito alto nível da tarefa a aprender, deixando a responsabilidade ao agente autónomo de lidar com os detalhes. A Aprendizagem por Reforço toma inspiração na natureza e na aprendizagem animal para modelar agentes que interagem com o seu ambiente de forma a escolherem as ações que aumentam a probabilidade de receberem recompensas e evitarem castigos. À medida que os agentes experimentam ações e observam os seus efeitos, ganham experiência e a partir dela derivam uma política. Isto é feito após cada observação do efeito de uma ação, ou após reunir conjuntos destas observações. Esta última alternativa, também chamada Aprendizagem por Reforço Batch, tem sido usada em aplicações reais com resultados promissores. Esta tese explora o uso de Aprendizagem por Reforço Batch para a aprendizagem de comportamentos para futebol robótico, tais como driblar a bola e receber um passe. Os resultados presentes neste documento foram obtidos de experiências realizadas com o simulador da equipa CAMBADA, assim como com os seus robôs.

Mestrado em Engenharia de Computadores e Telemática
No desenvolvimento de um agente inteligente e necess ario criar um conjunto de comportamentos, mais ou menos complexos, para que o agente possa escolher o que achar mais adequado para utilizar a cada instante. Comportamentos simples podem ser facilmente programados \ a m~ao", mas, a medida que se tentam criar comportamentos mais complexos esta tarefa pode tornar-se invi avel. Isto pode acontecer, por exemplo, em casos onde o espa co de estados, o espa co de a c~oes e/ou o tempo tomam valores cont nuos. E esse o caso no futebol rob otico, onde os rob^os se movem num espa co cont nuo, com velocidades e em tempo cont nuos. A aprendizagem por refor co permite que seja o agente a aprender um comportamento atrav es da sua experi^encia ao interagir com o mundo. Esta t ecnica baseia-se num mecanismo que ocorre na natureza, uma vez que imita a forma como os animais aprendem, mais concretamente, observando o estado do mundo, tomando uma a c~ao e observando as consequ^encias dessa a c~ao. A longo prazo, e com base nas consequ^encias das a c~oes tomadas, o animal aprende se, nessas circunst^ancias, a sequ^encia de a c~oes que o levaram a esse ponto e boa e pode ser repetida ou n~ao. Para que o agente aprenda da mesma forma, e preciso que consiga percecionar o valor que as suas a c~oes t^em a longo prazo. Para isso, e-lhe dada uma recompensa ou um castigo quando faz uma a c~ao desejada ou indesejada, respetivamente. Comportamentos aprendidos podem ser usados em situa c~oes em que e invi avel escrev^e-los a m~ao, ou para criar comportamentos com melhor desempenho uma vez que o agente consegue derivar fun c~oes complexas que descrevam melhor a solu c~ao do problema. No contexto desta tese foram desenvolvidos 3 comportamentos no contexto da equipa de futebol rob otico CAMBADA da Univeridade de Aveiro. O primeiro comportamento, o mais simples, consistiu em fazer o rob^o rodar sobre si pr oprio at e estar virado para uma dada orienta c~ao absoluta. O segundo permitia que o rob^o, com a bola na sua posse, a driblasse numa dire c~ao desejada. Por m, o terceiro comportamento permitiu que o rob^o aprendesse a ajustar a sua posi c~ao para receber uma bola que pode vir com mais ou menos velocidade e descentrada em rela c~ao ao receptor. Os resultados das compara c~oes feitas com os comportamentos desenvolvidos a m~ao que j a existiam na CAMBADA, mostram que comportamentos aprendidos conseguem ser mais e cientes e obter melhores resultados do que os explicitamente programados.
While developing an intelligent agent, one needs to create a set of behaviors, more or less complex, to allow the agent to choose the one it believes to be appropriate at each instant. Simple behaviors can easily be developed by hand, but, as we try to create more complex ones, this becomes impracticable. This complexity may arise, for example, when the state space, the action space and/or the time take continuous values. This is the case of robotic soccer where the robots move in a continuous space, at continuous velocities and in continuous time. Reinforcement learning enables the agent to learn behaviors by itself by experiencing and interacting with the world. This technique is based on a mechanism which happens in nature, since it mimics the way animals learn, more precisely, observing the world state, taking an action and then observe the consequences of that action. In the long run, and based on the consequences of the actions taken, the animal learned if, in those circumstances, the sequence of actions which led it to that state is good and may be repeated or not. To make the agent learn in this way, it must understand the value of its actions in the long run. In order to do that, it is given a reward or a punishment for doing a desired or undesired action, respectively. Learned behaviors can be used in cases where they are too complex to be written by hand, or to create behaviors that can perform better than the hand-coded ones, since the agent can derive complex functions that better describe a solution for the given problem. During this thesis, 3 behaviors were developed in the context of the robotic soccer CAMBADA team from University of Aveiro. The rst behavior, the most simple, made the robot rotate about itself until it had turned to a given absolute orientation. The second one, allowed a robot that possessed the ball to dribble it in a desired direction. Lastly, the third behavior allowed the robot to learn to adjust its position to receive a ball. The ball can come at a high or low speed and may not be centered in relation to the receiver. The results of comparing the learned behaviors to the already existing handcoded ones showed that the learned behaviors were more e cient and obtained better results.

In recent times there has been a substantial increase in research interest of soccer due to an increase of availability of soccer statistics data. With the help of data provider firms, access to historical soccer data becomes more simple and as a result data scientists started researching in the field. In this thesis, we develop prediction models that could be applied by data scientists and other soccer stakeholders. As a case study, we run several machine learning algorithms on historical data from five major European leagues and make a comparison. The study is built upon the idea of investigating different approaches that could be used to simplify the models while maintaining the correctness and the robustness of the models. Such approaches include feature selection and conversion of regression prediction problems to binary classification problems. Furthermore, a literature review study did not reveal research attempts about the use of a generalization of binary classification predictions that applies different target class upper boundaries other than 50% frequency binning. Thus, this thesis investigated the effects of such generalization against simplicity and performance of such models. We aimed to extend the traditional discretization of classes with equal frequency binning function which is standard for converting regression problems into the binary classification in many applications. Furthermore, we ought to establish important players’ features in individual leagues that could help team managers to have cost-efficient transferring strategies. The approach of selecting those features was achieved successfully by the application of wrapper and filter algorithms. Both methods turned out to be useful algorithms as the time taken to build the models was minimal, and the models were able to make good predictions. Furthermore, we noticed different features matter for different leagues. Therefore, in accessing the performance of players, such consideration should be kept in mind. Different machine learning algorithms were found to behave differently under different conditions. How-ever, Naïve Bayes was determined to be the best-fit in most cases. Moreover, the results suggest that it is possible to generalize binary classification problems and maintain the performance to a reasonable extent. But, it should be observed that the early stages of generalization of binary classification models involve a tedious work of training datasets, and that fact should be a tradeoff when thinking to use this approach.

Orientador: João Batista Freire
Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Educação Fisica
Made available in DSpace on 2018-07-26T05:35:10Z (GMT). No. of bitstreams: 1 Scaglia_AlcidesJose_M.pdf: 7598948 bytes, checksum: f657618eda6f53ca511fa3de2a2ba785 (MD5) Previous issue date: 1999
Resumo: Tendo por objetivo discutir e analisar o processo de ensino-aprendizagem do futebol, nosso estudo, partindo da hipótese de que não se ensina mais futebol da maneira como se aprendia, procurou levantar junto a ex-jogadores de futebol que hoje ministram aulas em escolinhas, o processo que lhes ensinaram a jogar futebol, para, na seqüência, compará-lo às suas respectivas práticas profissionais. Para a realização desta pesquisa de campo, a sistematização metodológica, tanto na coleta quanto na análise dos dados, seguiu os procedimentos da técnica de análise de conteúdos. Porém, para que realizássemos as análises inferenciais comparativas com segurança e coerência científica, fez-se necessário que construíssemos um corpo teórico consistente. Este corpo teórico, num primeiro momento, levantou, junto à história do futebol, suas origens, descrevendo toda a evolução dos antigos jogos de bola com os pés que culminou com o surgimento desse esporte na Inglaterra. Depois, procurou-se levantar hipóteses de que o futebol continuou sua evolução no Brasil através das inúmeras brincadeiras de bola com os pés realizadas pelas crianças ao longo da infância. O final desse resgate histórico evidenciou o surgimento das escolinhas de futebol, caracterizando-as como os locais que vieram substituir a rua e os campinhos, principalmente em grandes centros como Campinas, no desenrolar da iniciação. Num segundo momento, tecemos considerações a respeito da necessidade de se relacionar a pedagogia com o futebol, almejando destacar pontos que deveriam ser levados em consideração ao se desenvolver um trabalho de iniciação ao futebol, caracterizando as escolinhas como instituições de ensino não-formal que, como tal, não se vê desprovida de responsabilidades maiores que ensinar apenas o futebol. Finalizando o corpo do texto, foi desenvolvida uma pesquisa bibliográfica que discutiu os vários estudos teórico-práticos publicados com a finalidade de ensinar futebol para crianças, ousando chamar alguns de tradicionais e outros de inovadores, ao se apontar suas características e particularidades. Ao final da dissertação pudemos confirmar as nossas hipóteses, diagnosticando que a pedagogia com que se aprendia futebol não é a mesma que ensina hoje nas escolinhas
Abstract: Having the objective of discussing and analyzing the soccer teaching ¿ leaming process, our study started with the hypothesis that nowadays soccer is not taught the same way it was leamed before. We tried to leam from fonner soccer players (who in the present teach at soccer schools) the process in which they were taught. After this, we could compare the data with their present professional activity. To get the results of this field research, the methodological systematization for the data collection and analysis followed the proceedings of the analysis of content technique. Therefore, for us to acquire safe comparative conclusion analysis and scientific consistency it was necessary that we built a consistent theoretical body. This one, in a first moment searched the soccer history, its origins, describing the evolution ofthe ancient ball games played with the feet from which emerged the soccer game in England. After this, we tried to bring a hypothesis that the soccer game continued its evolution in Brazil throughout several ball games the children would play with their feet during childhood. The end of this historical research made evident the appearance of the soccer schools, pointing out that these places once replaced the street soccer games and the neighborhood soccer fields. These were places where the children would start leaming soccer especially in the big cities like Campinas. In a second moment, there were considerations about the need of relating pedagogy with soccer, with the expectation of high1ighting some points that should be considered when developing a work for initiation in soccer. We distinguished the soccer schools as non fonnal educational institutions. These institutions, although non fonnal, have the responsibility of teaching more than soccer. Concludingthe text, we developed bibliographical research in which we discussed several theoretical-practical studies published with the purpose of teaching soccer to children. When pointing out their characteristics and peculiarities we took the risk of calling some traditional and some innovators. At the end of the dissertation, we could confirm some hypotheses, concluding that the pedagogy which the soccer players first learned is not the same used nowadays to teach at the soccer schools
Mestrado
Mestre em Educação Física

This thesis aims at providing a cross-cultural study of how football museums represent and construct identities, both collective and personal. The research is based on a multi-sited ethnography at selected football museums in the UK, Germany, and China, employing participant observation, photographic recording and online research methods. This investigation sharpens an anthropological awareness of constructions of multiple layered identities by examining football museums' exhibiting practices and activity programmes, as well as their built environments and cultural settings. The research also offers a perspective on museum visitors, who consume football museums with diverse personal and collective identity claims. Looking into the largely under-explored terrain of football museums, this research joins continuing anthropological efforts to understand identity work while also exploring continuing tensions inherent in a marriage between museums and football. The thesis contributes to the research field of football/sports museums with an ethnographic emphasis and a cross-cultural range.