BEIJING, CHINA / gridsum.com / September 7, 2017 – Gridsum Holding Inc. (“Gridsum” or the “Company”) (NASDAQ:GSUM), a leading provider of cloud-based big-data analytics and artificial intelligence solutions in China, held the final contest of its Third Annual Gridsum Hackathon successfully in Beijing on August 19, with 16 cross-region, cross-division finalist teams. “Gridsum Deep Vision Dissector (GDVD)” designed by the employees from the R&D and Enterprise Business departments won the first prize, followed by the Intelligent Legal Case Result Prediction Engine, and Automatic Anomaly Detection in Digital Advertising. Gridsum’s DNA of world-leading innovation was fully demonstrated in the Hackathon.
The Gridsum Hackathon, now in its third year, promotes a defined culture of cutting-edge innovation. Teams compete in a fun, competitive, intense and genial environment, to push the envelope of what is possible with global “best of breed” big data and enterprise-focused AI technology. These teams develop technologies and applications directed toward adding value for enterprises and other organizations and towards enhancing Gridsum’s suite of SaaS solutions.
38 teams with 128 Gridsum employees competed in the Gridsum Hackathon this year. The teams have a hybrid composition, across company divisions and functions, with members from R&D, sales, HR and across business lines including Enterprise, Legal Services, New Media and Government.
Gridsum CEO Guosheng Qi spoke at the opening and encouraged contestants to think about “effective communication” of the projects focus and benefits, “high impact deployment” of the winning technologies into Gridsum’s solutions and “seamless integration” with Gridsum’s SaaS solution suite for clients.
Gridsum Deep Vision Dissector (GDVD) was the first prize winner this year. Powered by Gridsum Prophet, Gridsum’s Enterprise AI engine, GDVD recognizes products, company and brand logos, and other specific objects embedded in video streams and images with great accuracy. It uses deep residual neural networks to better extract image characteristics and create image recognition models through deep learning. This capability enables Gridsum’s industry leading Marketing Dissector suite of products and solutions to take advantage of visual data which has applications in all aspects of digital marketing, such as advertising monitoring and evaluation, brand protection, customer sentiment analysis, and effective marketing budget allocation across all digital channels and media platforms.
The Intelligent Legal Case Result Prediction Engine drives predictive analysis for criminal and civil cases across verdict outcome and expected penalty.
Automatic Anomaly Detection in Digital Advertising further adds to Gridsum’s click fraud detection canon with further applications of the technology anticipated across the company’s product development pipeline.