Federated-Learning-Based Method and System Secures the Data Privacy on Personal Mobile Devices

Bairong Inc., a leading independent AI-powered technology platform in China, has been making efforts to develop new technologies to better protect personal privacy. Its new patent based on federated learning can effectively protect privacy by restricting certain personal data from being transmitted from users’ mobile devices.

Marketing Technology News: Bairong’s New Precision Marketing Solution Empowers Financial Institutions to Make Data-Driven…

The patent – “A credit anti-fraud prediction method and system for mobile devices based on federated learning” – can be applied as an important data processing and analysis solution to protect private personal information. With individual private data not leaving users’ mobile devices, the solution can realize joint modeling between individual users and commercial institutions, so as to avoid leakage of personal information from the source and secondary leakage under technical analysis.

“Data security relies on advanced privacy computing technologies,” said Shaofeng Zhang, Founder and CEO of Bairong. “It is crucial to lay a solid foundation for privacy protection with patented technologies before commercial applications are widely deployed. We have long recognized the importance of data security for the digital economy, and have been developing innovative technologies in the field of data security and privacy computing.”

In a common scenario of financial data collection and analysis, if users want to borrow money online through their mobile devices, they first need to send their personal information, including app list, contact information, geographical location, and equipment identity number, to the server of financial institutions during the application process. The privacy and security of personal information may be at risk in this process.

With Bairong’s innovative patent, users no longer need to send their personal information from their mobile devices. The new solution can directly develop a federal learning model between users’ mobile devices and financial institutions, which can avoid leakage of user information from the source. While solving the problem of data security, more information on mobile devices can be leveraged to improve the accuracy of the prediction model.

Bairong’s patented technology has already been adopted by some financial institutions. For a key FSP client, Bairong’s federated learning technology was applied to upgrade its credit evaluation model. By ensuring security of data privacy and compliance, Bairong broke through “isolated island” limitations and integrated data from multiple parties, which increased the client’s model evaluation index K-S, directly improved the asset business efficiency with AI enabling transformation, and promoted cross operations with wealth management business.

China’s long-term blueprint for economic development has prioritized data security to include protection of trade secrets and personal privacy, legislation of data security and personal information protection, and security of data resources throughout their life cycle.

Bairong’s patent on privacy protection is a pioneering technology in the industry, which will provide an innovative solution to data security and business development in the future. In recent years, Bairong has increased investment in R&D, and developed advanced technologies in artificial intelligence, cloud computing, blockchain, and privacy computing.

Marketing Technology News: MarTech Interview with Matt Ramerman, President at Sinch for Marketing