An AI-based betting anomaly detection system has been developed to ensure fairness in sports and prevent illegal gambling. The study utilizes machine learning models to detect anomalies in sports match-fixing based on betting odds. Five models, including logistic regression, random forest, support vector machine, k-nearest neighbor classification, and an ensemble model, were used to distinguish between normal and abnormal matches by learning their patterns using sports betting odds data. The database was constructed based on world football league match betting data from 12 betting companies, offering a vast collection of data on players, teams, game schedules, and league rankings for football matches.
An abnormal match detection model was developed based on the analysis results of each model, using match result dividend data. Real-time match data was collected and the five models were applied to construct a system capable of detecting match-fixing in real-time. The RF, KNN, and ensemble models recorded high accuracy rates of over 92%, while the LR and SVM models were approximately 80% accurate. The study aims to contribute to the eradication of match-fixing in sports by leveraging AI technology and thorough data analysis to ensure integrity within sports.
Sports events are governed by rules and fair competition, but match-fixing poses a threat to the integrity of sports competitions. Efforts to ensure fairness in sports are ongoing, as match-fixing can lead to substantial negative consequences for the sports industry. Anomalies in sports play a crucial role in detecting match-fixing, as they can manifest in various forms such as unusual betting patterns, unexpected performance fluctuations, or suspicious player behaviors. By analyzing these anomalies, instances of manipulation can be uncovered, and appropriate action can be taken to maintain the fairness of sports competitions.
Market risks associated with match-fixing can have significant negative consequences, threatening the integrity of sports and causing fans to lose interest. Detecting behaviors of athletes and those involved in match-fixing is crucial, and various studies have been conducted to develop detection methods using player behavior patterns and betting odds data. Betting odds are a key factor in detecting match-fixing, as they can provide insights into the potential manipulation of match outcomes. Continuous efforts are being made to build systems for detecting abnormal signals in sports to eliminate cheating and promote fair play.
The study utilized a sports betting database to detect anomalies and prevent match-fixing through data analysis. The database included data on sports teams, match results, and betting odds, which were used to develop a match-fixing detection model. By analyzing betting odds data and leveraging machine learning models, the study aimed to identify abnormal matches and contribute to the eradication of match-fixing in sports. The developed system showed promising results in classifying normal and abnormal matches, with high accuracy rates recorded for certain models.
📰 Related Articles
- Sports Betting Scandals Prompt Calls for Industry Integrity
- How NBA Addresses Integrity Risks in Sports Betting Prediction Markets
- Exploring AI Sports Betting Bots: Efficiency, Risks, and Strategies
- AI Transforms Sports Betting with Data-Driven Predictions
- AI Transforms Sports Betting with Data-Driven Predictions