Evaluation Of Competitive Sports Team Strength Based On Hadoop

With the development of the Internet and people's increasing attention to various competitive events,top competitions such as the U.S.professional basketball league,the UEFA Champions League,and the FIFA World Cup are among the largest and most successful leagues in the world today.It not only attracted the world's best athletes to gather here,but also led the development direction of the sport.Not only the entertainment stars,but also the sports circles have enthusiastic players.The admiration of the stars,the passionate confrontation of the athletes,and the successful commercial promotion have made it more and more popular with people all over the world.The major sports betting companies and large-scale sports websites seized this opportunity to conduct betting and statistics on the games,and provided the historical records of the major groups and ranked the team's winning and losing.The followers of the competition often evaluate the strength of the team according to the ranking of the winners and losers,or assess the team strength based on their intuition and preferences.However,the actual group strength is difficult to withdraw from one or two games.Not only must the team's winning rate be considered but also the strength of each other in the team competition.It must be reflected after a long season of competition.The ordinary followers are even more.It is difficult to accurately predict the outcome of the next round of games so that gaming companies can continue to operate without losses.Since the winning percentage does not accurately reflect the strength of the group,in the era of big data,we should conduct a scientific and reasonable assessment based on the existing historical competition data.In order to more accurately guess the winning percentage of various teams in order to more accurately guess the above questions,this paper selects the Page Rank algorithm that is used to evaluate the ranking of the pages to evaluate the team strength,and selects the actual game data of the NBA 30 team regular season as the experimental data set.After expounding the principle of Page Rank algorithm,the method of assessing the strength of competitive groups was realized by using PageRank algorithm.According tothe experimental results,it was verified that Page Rank algorithm had better advantages.Then it was pointed out that Page Rank algorithm was used to evaluate the insufficiency of team strength,and on this basis,it was proposed.Based on the PageRank algorithm's improved algorithm,the Sport Rank algorithm,Sport Rank inherits the advantages of the Page Rank algorithm.At the same time,it takes into account two key factors: the strength of the opponents in the historical game and the score difference between the teams in the game,compared with the traditional assessment team.The method is more comprehensive.Finally,the Hadoop big data processing platform is introduced.The Page Rank algorithm and the SportRank algorithm are used to calculate the massive data through the Hadoop platform.The comparison of the experimental results proves that the Sport Rank is more accurate and reliable.