How to create a high-precision risk control model through Graph+AI?

Introduction to Cloud Map Intelligent Platform has helped customers in banking, insurance and other fields in the financial industry to build financial risk control, product recommendation, recurring guarantee detection, abnormal indicator monitoring, illegal group mining and other scenarios. By penetrating industry application scenarios, Help customers make accurate decisions based on multidimensional data.

\>>Conference Portal : https://yqh.aliyun.com/live/gdb2021

"Gartner 2021 Top Ten Data and Analysis Trends" pointed out that graph technology makes everything relevant. It is predicted that by 2025, the proportion of graph technology in data and analysis innovation will rise from 10% in 2021 to 80%. The technology will facilitate rapid decision-making throughout the enterprise organization. From the perspective of the financial industry, driven by policies such as the "FinTech Development Plan (2019-2021)" issued by the People's Bank of China, the construction of a financial knowledge graph to make decisions based on multi-dimensional data sources can effectively drive financial institutions to reduce costs Synergistic.

The graph database GDB is a graph database product independently developed by Alibaba Cloud. It has experienced a wealth of application scenarios within the Alibaba Group and has a wealth of best practices. Graph database GDB entered the Forrester graph data platform competitor quadrant in 2020, and it was also the first time that a domestic graph database product was selected. Alibaba Cloud Graph Database GDB not only meets the requirements of high reliability and high performance, but also takes into account the characteristics of low cost. The cost of product use and operation and maintenance is only 40% of foreign graph database products. We sink AI capabilities such as automatic feature engineering and automatic machine learning into the graph database engine to form an Alibaba Cloud graph intelligence platform, making the entire graph model construction, analysis, and release process naturally coherent. In the financial industry, Alibaba Cloud Atlas’ intelligent platform has helped customers in banking, insurance and other fields to build financial risk control, product recommendation, cyclic guarantee detection, abnormal indicator monitoring, illegal group mining and other scenarios. By penetrating industry application scenarios, helping customers based on multi-dimensional Data makes precise decisions.

The traditional financial risk control model can gather the attribute characteristic information of each data source, but it is difficult to mine the deep association relationship between the data sources. To deeply and quickly mine the associated features of massive data, it will face very big technical challenges. The significance of graph technology is to upgrade information, while the significance of machine learning technology is to summarize data laws. Through the graph representation learning technology, the topological information features in the financial knowledge graph are extracted, and through the graph automatic feature engineering module, the features are automatically constructed as the input conditions of the risk control model to participate in the model training. Through the automatic machine learning module, it helps financial institutions to select, debug, and integrate various machine learning models to achieve a higher-precision risk control model.

Huarui Bank officially introduced Alibaba Cloud Map database GDB in 2020 to further improve risk identification capabilities through in-depth correlation analysis of data assets. By building a set of enterprise-level graph analysis platform, intelligent risk management and control of businesses such as smart supply chain and air travel consumer loans have been realized. Through the automatic machine learning component integrated with Alibaba Cloud Map database GDB, Huarui Bank has significantly reduced the risk control model development cycle, and has detected 6 fraud groups in practice so far, effectively preventing and controlling business risks.

The application of graph database can mine the in-depth relationship between data sources in highly related data, and enhance the information through understanding and analyzing graphs, thereby helping enterprises to gain insights, which will become the core competitiveness of enterprises in the future. We will also continue to improve our graph database products and services, and explore the real needs of users, in order to help more companies and developers gain insight and competitive advantage.

Copyright statement: content of this article is contributed spontaneously by Alibaba Cloud real-name registered users. The copyright belongs to the original author. The Alibaba Cloud Developer Community does not own its copyright and does not assume corresponding legal responsibilities. For specific rules, please refer to the "Alibaba Cloud Developer Community User Service Agreement" and the "Alibaba Cloud Developer Community Intellectual Property Protection Guidelines". If you find suspected plagiarism in this community, fill in the infringement complaint form to report it. Once verified, the community will immediately delete the suspected infringing content.