DeepBlue, not only an enterprise, also a platform for the realisation of dreams

DeepBlue Technology emerges champion at top security informatics competition organized by IEEE.


DeepBlue Technology emerged champions at this year’s IEEE ISI 2019 International Big Data Analysis Competition (IWC 2019), as well as a third-place finish across two different tracks: Company Investment Value Evaluation and Lawsuit Category Prediction respectively. Deepblue’s Automated Machine Learning (AutoML) solutions scored a large leading edge of over 0.1 over the second-placed team, with a score of 3.2585.

The annual International Conference on Intelligence and Security Informatics (IEEE ISI) is the flagship conference in the field of security informatics. In the past 16 years, the event has evolved from the traditional field of intelligence and security to multi-field joint research and innovation. This year, the 17th IEEE ISI Conference was held in Shenzhen, China, from July 1 to 3.

IWC 2019
IWC 2019

The IWC 2019 competition was launched in order to promote the development of AI analysis industry and provide a good platform for academic exchanges and technical discussions, and is open to universities, research institutes, enterprises and governments all over the world. This year, over a thousand participants joined the competition. More than 300 teams from China along with other teams from the United States, Pakistan, the United Kingdom, Germany and seven other countries participated in the contest.

For this competition, the DeepBlue used an AutoML solution which can automatically interpret higher-order features by extracting information from various dimensions of the enterprise, such as mining complex relationships among business fields and providing a novel and more accurate solution for enterprise value evaluation. According to a team spokesperson: "Through this competition,we not only obtained a very good result, but also verified the stability and feasibility of the system and underlying technologies. In the future, through continuous optimization and improvement, we can design more features for different business scenarios, such as data mining, information extraction, network public opinion and other fields. We can then take AutoML technology into the implementation phase and use it in everyday products.”

In addition to saving a lot of manpower, material and financial resources, AutoML technology can also be used to build a machine learning system faster and safer than most algorithmic engineers. At present, AutoML has been widely used in business scenarios such as precision marketing, financial wind control, automatic driving, disease prediction amongst others.