Context AI in Sports Investigations

Challenge

The organization’s Anti-Match-Fixing Unit relied on a basic case management solution that lacked advanced investigative capabilities such as search, translation, entity resolution, link charting, visual queries, and automated summaries. These limitations hindered their ability to efficiently manage suspicious match alerts and related intelligence.

Key challenges included:

• Inefficient handling of diverse data types and formats – including betting records and OSINT. While open-source data collection remained a manual process, integrating this data into a comprehensive investigative picture was made possible with DataWalk.
• Lack of a common ontology, making it difficult to visualize a single source of truth that connects various entities, such as players, teams, clubs, and reports of suspicious activities.
• Multilingual complexities – Even player names created difficulties (e.g., Juri, Juriy, and Yuriy referring to the same individual). Beyond translation, effective entity extraction and resolution (entity matching) posed a significant challenge.
• Time-consuming manual processes – Analysts had to manually gather and connect information from multiple sources, including betting records, match events, locations, and incidents.
• Rigid case management tools – While existing solutions provided basic integration, they were cumbersome when incorporating new types of OSINT-derived information.
• Lack of flexibility in adapting to new threats – The existing system’s inability to handle evolving data types limited its adaptability and responsiveness to emerging risks.

Solution

The organization embarked on a transformative journey by implementing DataWalk’s AI-Powered Investigative Software Platform with Graph Technology. This solution empowered intelligence teams by analyzing both internal data and external sources to:

• Build a reliable ontology using a knowledge graph that links people, events, organizations, and locations into a simple, intuitive structure—including individuals with multiple roles (e.g., a player transitioning to a coach or referee, with these changes recorded over time).
• Accelerate data linking from unstructured sources by leveraging AI-driven language translation, entity recognition, and alias resolution (e.g., recognizing that “Jordan” may also be referred to as “MJ,”).
• Automatically detect duplicate records, improving data quality and reliability.
• Support preventive measures against match-fixing by identifying new patterns and risks using no-code visual queries.

Implementation & Results

The adoption of DataWalk transformed the organization’s approach to investigations of match integrity, delivering significant improvements:

• Advanced Entity Extraction and Resolution – The platform’s AI-driven capabilities streamlined the processing of multilingual, unstructured data.
• Enhanced Search Capabilities – Fuzzy search eliminated duplicate data, enabling centralized intelligence management.
• Rapid Hypothesis Testing – Investigators could explore new leads through no-code querying, scoring, and alerting mechanisms.
• Comprehensive Visualization – DataWalk’s tools provided clear, interactive representations of complex networks involving players, teams, and their affiliates, dramatically improving investigative efficiency.
• Automated Relationship Discovery – The FindPath feature enabled the automatic detection of indirect connections between entities, such as identifying patterns where two players had participated in multiple suspicious matches.

Furthermore, the integration of DataWalk with the organization’s new case management tool streamlined workflows, enabling seamless data analysis. This unified investigative environment allowed for detailed insights into networks of players, coaches, teams, and clubs.

While Clue served as a case management and data integration tool, DataWalk significantly enhanced its openness to new data sources, including manually uploaded files and third-party providers like Stats Perform. Additionally, DataWalk was fully equipped to incorporate betting statistics, further enriching investigative efforts.

Conclusion

By implementing DataWalk, the organization set a new benchmark in the fight against match-fixing. The integration of advanced analytics, AI, and graph-based investigative tools has strengthened its ability to uphold the integrity of the sport. This case study underscores how cutting-edge technology can empower global sports organizations to maintain fairness and transparency in competitive disciplines.