Based on digital technologies such as data asset management, MPC multi-party security computing, federated learning, blockchain, the privacy computing platform AISWare MPC realizes data availability with invisibility, connects enterprises and industry data islands, helps enterprises build trusted data circulation and transactions, activates the value of data elements, and releases huge dividends and powers brought by data elements.
Achieving data availability with invisibility and building trusted privacy computing system
Revitalizing enterprise data assets and improving trusted data circulation and transactions
Integration and analysis of data elements empower digital innovation in the industry
The "1 + X " architecture of private data computing realizes decoupling of algorithms and computing power, and decoupling of data and algorithms, and supports one-touch integration of heterogeneous algorithms; It has the self-research capability of core algorithms and can empower opening up and interconnection with multi-technologies.
Encapsulating the general operator components in the industry and providing workflow component arrangement and flexible component configuration. Realizing the visualization and drag-and-drop configuration of the privacy computing process and creating an efficient out-of-the-box capability.
Multi-field logical scenarios have been accumulated to meet the requirements of different scenarios, quickly empowering vertical industry applications and accelerating enterprises to build an aggregated data ecology.
Interconnection of heterogeneous platforms
Activating data values
Empowering industry innovation
Helping enterprises form an industrial ecology
To meet the strategic requirements of operators' big data conducting external empowerment and the regulatory requirements of national data security protection laws and regulations, AsiaInfo built a privacy computing platform for an operator group company that adopted centralized construction and provincial operation mode. This project has completed the introduction of federated learning, multi-party security computing, and other technologies, built basic data security integration, query, computing, and modeling capabilities, formed an enterprise-level security capability platform, and formulated a multi-party security computing technical scheme for service operators. The construction of the operator's privacy computing platform can significantly enhance the application value of the operator's massive data and bring new value to the operator's revenue based on the operator's big data advantages.
The operator possesses many users' attribute and behavior data, such as online logs, APP records, and call records. However, the data cannot be empowered for external business due to data partition. Based on the encapsulation of the underlying technology for the multi-party security computing system, a mobile privacy computing platform combines the federated framework and AI algorithm to create an efficient and secure system architecture, interacting with the centralized big data service management and control system to obtain data assets, encapsulate the model and register and convey the computing results, supporting several data service scenarios such as administrative service, medical care, finance, retail.