PDF(Pubmed)
Abstract：
This study develops a solution to sports match-fixing using various machine-learning models to detect match-fixing anomalies, based on betting odds. We use five models to distinguish between normal and abnormal matches: logistic regression (LR), random forest (RF), support vector machine (SVM), the k-nearest neighbor (KNN) classification, and the ensemble model-a model optimized from the previous four. The models classify normal and abnormal matches by learning their patterns using sports betting odds data. The database was developed based on the world football league match betting data of 12 betting companies, which offered a vast collection of data on players, teams, game schedules, and league rankings for football matches. We develop an abnormal match detection model based on the data analysis results of each model, using the match result dividend data. We then use data from real-time matches and apply the five models to construct a system capable of detecting match-fixing in real time. The RF, KNN, and ensemble models recorded a high accuracy, over 92%, whereas the LR and SVM models were approximately 80% accurate. In comparison, previous studies have used a single model to examine football match betting odds data, with an accuracy of 70-80%.
摘要：
这项研究开发了一种解决方案，使用各种机器学习模型来检测体育比赛固定异常，基于投注赔率。我们使用五个模型来区分正常匹配和异常匹配：逻辑回归(LR)，随机森林(RF)，支持向量机(SVM)，k-最近邻(KNN)分类，和集成模型-从前四个优化的模型。模型通过使用体育博彩赔率数据学习其模式来对正常和异常比赛进行分类。该数据库是根据12家博彩公司的世界足球联赛比赛博彩数据开发的，提供了大量关于玩家的数据，团队,游戏时间表,和足球比赛的联赛排名。根据各模型的数据分析结果,建立异常匹配检测模型,使用匹配结果红利数据。然后，我们使用来自实时匹配的数据，并应用这五个模型来构建一个能够实时检测匹配修复的系统。RF,KNN,合奏模型记录了很高的准确性，92%以上,而LR和SVM模型的准确率约为80%。相比之下,以前的研究使用单一模型来检查足球比赛投注赔率数据，准确率为70-80%。
