China Mobile Group Jiangxi Co., Ltd.
Next-Generation Telecom Network Fraud Prevention, Deterrence, and Traceability System Based on Graph Databases
“Preventing and combating telecom fraud inherently aligns with the core strengths of graph databases, which excel at modeling interconnected entities and intuitively representing relationships within data. The efficiency of graph databases in processing relationship data is 2 to 3 orders of magnitude higher than that of relational databases. In terms of processing relationship data, graph databases demonstrate efficiency that is 2 to 3 orders of magnitude higher than that of relational databases”.
Deputy General Manager of the Network Department at Ganzhou Branch of Jiangxi Mobile - Xie Wenfeng
Customer Profile
China Mobile Communications Group Jiangxi Co., Ltd. (hereinafter referred to as “Jiangxi Mobile”) is a subsidiary of China Mobile Communications Corporation. It leads the industry in terms of user scale and revenue, and has achieved outstanding performance within the China Mobile Group for four consecutive years.
Business Challenges
China Mobile previously relied on a traditional Oracle relational database for its anti-fraud models, but has hit performance bottlenecks.
Slow Query Performance
The sheer volume of internal user data led to slow query response times. As the user base expanded, query latency increased exponentially.
Frequent System Crashes
Jiangxi Mobile generates over 100 million new user call records daily. The existing system, constrained by hardware limitations, often failed to return query results efficiently, leading to frequent system crashes
Complex Model Statements
Anti-fraud models required frequent iterations. However, these models were built using SQL, which necessitated multiple join operations across relational databases. This resulted in overly complex and cumbersome statements, significantly reducing readability and maintainability.
solution
Jiangxi Mobile established a pilot team for graph database anti-fraud efforts and built a next-generation anti-fraud system with Galaxybase Graph Database.
Conversion
Transformed massive legacy relational data structures into graph data structures to create an “anti-fraud graph”.
Model
Migrated existing anti-fraud models to the graph database, while developing and continuously optimizing advanced graph-based anti-fraud models.
Mining
Leveraged graph database capabilities to deeply mine relationships between high risk numbers, trace and target fraud ring leaders, and neutralize fraud at its source
Architecture Diagram
Business Value
Jiangxi Mobile set up a graph database anti-fraud pilot team and built a new-generation anti-fraud system with Galaxybase Graph Database.
Speed
Detection and alerting have been dramatically accelerated. The time to identify fraudulent numbers and notify potential victims has been slashed from hours—under legacy systems—to minutes, enabling near real-time intervention.
Accuracy
By constructing a trusted user database leveraging the inherent relational patterns of legitimate users, the system reduced the graph model’s false positive rate by 30%. This enables highly precise fraud detection through continuously enriched knowledge graphs.
Effectiveness
The system uncovers hidden connections between fraudulent actors with high precision. By targeting ringleaders directly, it strikes at the root of telecom fraud—proactively dismantling organized networks.
Customer Benefits
This achievement has been piloted in Ganzhou City, Jiangxi Province. In the first month after application, telecom fraud cases at Ganzhou Mobile dropped by 69.23%, with an annual decrease of approximately 60%. The number of valid clues provided to public security authorities increased significantly: 279 valid fraud-related clues were submitted annually, leading to the dismantling of nearly 200 fraud rings and the prevention of over 200 potential fraud cases.
Create Link is the first domestic commercial graph database supplier with fully independent intellectual property rights, dedicated to providing world-class graph database products and graph intelligence services.
Business Consulting
400-882-6897
Pre-sales Consulting
0571-88013575、0571-88016275
Enterprise Email
partner@chuanglintech.com
Media Cooperation
Marketing@chuanglintech.com
Address
Room 605, Building 6, No.666, Zhenhua Road, Sandun Town, Xihu District, Hangzhou, Zhejiang, China
