Large-Scale Complex Heterogeneous Graph Data Intelligent Analysis Technology Wins 2022 CIEE Science and Technology Award

On February 21, 2022, the highest honor in the electronic information field, the 2022 China Institute of Electronics (CIEE) Science and Technology Award, was announced, with 17 technological achievements—including the "Large-Scale Complex Heterogeneous Graph Data Intelligent Analysis Technology and Scalable Application"—receiving the First Prize for Scientific Progress.

Graphs, as a mathematical concept, excel at representing relationships between entities, and graph technologies that store, compute, and model graph data are crucial for handling massive, heterogeneous, and dynamic big‑data characteristics, becoming a strategic focal point in global technology competition.

In the highlighted project, Ant Group and Beijing University of Posts and Telecommunications jointly tackled core technologies, partnering with China Mobile, HaiZhi Star Map, and others to develop a full‑chain solution covering theory, industrial application, and system implementation. They built complex graph neural network models, created the trillion‑scale dynamic heterogeneous graph learning system AGL, and released the OpenHGNN platform, overcoming previous limitations in modeling capability, structural knowledge utilization, and massive data computation.

The project has produced 43 invention patents, 51 CCF‑A class papers, contributed to two standards, and its graph machine‑learning systems and tools have passed testing by the China Academy of Information and Communications Technology, with 30 functional and performance indicators leading the industry.

These technologies are now maturely applied in financial risk profiling, intelligent semantic recommendation, supply‑chain identification, and network traffic analysis. For example, NetEase’s "Great Goose" system raised small‑business loan approval rates from 30% to 80%, and the technology supports China Mobile’s recommendation services for hundreds of millions of users and power supply for the 2022 Beijing Winter Olympics venue, saving over 7 billion yuan for more than ten enterprises across finance, internet, communications, and power sectors.

The award, approved by the National Science and Technology Awards Office, selects top projects through a rigorous three‑round review by dozens of academicians, narrowing down from over 300 submissions to a select few for the First Prize.