GBase Strikes at Telecom Fraud | Empowering Hunan Mobile's Anti-Fraud Big Data Platform

With the rapid development of information communications, big data and other technologies, traditional crimes are accelerating their shift to digital and networked forms, leading to a surge in new types of crimes represented by telecom and online fraud. In recent years, Hunan Province has seen frequent telecom fraud cases due to factors like its geographical position, large population, massive migrant worker outflow, and historical migration to Yunnan for border support.

To combat telecom and online fraud, China Mobile Hunan Co., Ltd. has built a dedicated anti-fraud big data platform. Leveraging a technology architecture that combines GBase database with Hadoop, it creates an integrated batch-stream processing model, relying on capabilities such as clue discovery and traceability to effectively analyze and handle fraudulent IPs, domain names, and apps, and to improve the early warning and dissuasion mechanism for potential victims, thus striking back at telecom fraud.

The GBase distributed logical data warehouse GBase 8a MPP Cluster underpins the implementation of Hunan Mobile's anti-fraud big data platform. With outstanding query and analysis capabilities in massive data scenarios, it effectively integrates Hunan Mobile's robust cloud, network, edge, and terminal information resources, providing a self-controlled and reliable technology tool for combating telecom fraud.

By building the DWS (Data Warehouse Service) layer for data processing in the anti-fraud big data platform model using GBase 8a MPP, the system not only resolved the system latency caused by task queuing due to large volumes of Hadoop computing resources being occupied by 15-minute and 30-minute scheduling jobs, but also leveraged GBase 8a MPP's performance advantages in massive structured data to analyze complex anti-fraud models and then issue instructions to various disposal systems, thereby enhancing the overall early warning capability of the anti-fraud big data platform.

After the system went live, the time to handle risky phone numbers was reduced from 2 hours to 7 minutes, the accuracy of identifying major fraud methods increased to 98%, and over 500,000 dissuasions of potential victims were conducted, significantly strengthening the crackdown on telecom and online fraud. Following its deployment in the second half of 2023, user complaints related to fraud governance decreased by 50.6% compared to the first half of 2023.

The "Smart Net" is cast wide and lets nothing slip through. GBase database supports the launch of Hunan Mobile's anti-fraud big data platform, weaving a dense technological net to safeguard people's money and offering new approaches to prevent online telecom fraud.

It is foreseeable that in the future, self-controlled and digital-intelligent products represented by GBase database will play a key role in network and digital security, using cutting-edge technology to help build a long-term mechanism for anti-fraud governance and safeguard every citizen's money.