The 16th ChinaJoy kicked off in Shanghai, China on August 3, 2018. The Global Game Industry Summit was held concurrently. David Ting, Vice President of Engineering from DataVisor, gave a speech titled “How AI Technology Can Improve the ROIs of Acquiring New Users” where he discussed the fake app install problems and introduced DataVisor’s fraud detection technology for the mobile gaming industry.
In 2018, fraudsters made a profit of over one billion dollars in marketing expenses.
According to statistics from TUNE, the world’s leading marketing data analysis platform, tens of millions of dollars are spent on the promotion of mobile games every year. With the rapid growth of revenue and users, gaming companies are increasingly targeted by fraudsters. In 2018 alone, more than $1 billion in marketing budget were wasted because of fraudulent activities. In general, the fraud rate for app promotion and installation is 10%-20%, but some ad channels have a fraud rate of over 50%. In the face of growing fraud, how to identify fraudulent ad channels has become a critical and difficult problem for gaming companies.
“Defining the impact and type of fraud is the first step in developing anti-fraud programs and improving ROI,” David Ting stated in his speech. Techniques leveraged by fraudsters are becoming more complicated and the behaviors of fraudsters are increasingly similar to those of real users, which greatly increases the difficulty for game companies to effectively differentiate real users from fraudulent users. Naive fraud attacks are highly dependent on low-cost labor, such as manual installation and mobile device simulators. More sophisticated fraudsters masquerade login activity from different devices using machines and scripts. The most advanced fraudsters are able to leverage advanced techniques to deceive existing detection rules, mimic the behavior of real users, and generate fake installations and activity.
DataVisor leverages AI to clear false traffic.
DataVisor was founded in Silicon Valley, and 65% of its team members are experts in machine learning and security from world-renowned schools such as Carnegie Mellon and Stanford. DataVisor’s patented and proprietary unsupervised machine learning fraud detection algorithm, superior to traditional blacklists, rules systems, or supervised machine learning detection methods, can automatically process all users’ accounts and events without labels or training data. Through its ability to cluster and correlate events in real-time, it effectively identifies hidden correlations and suspicious events to detect entire fraud groups. DataVisor’s algorithm greatly improves detection coverage and accuracy to help advertisers target real users.
DataVisor’s fraud detection solution has protected over 4 billion accounts for the world’s gaming, social, e-commerce, and financial companies. DataVisor currently works with well-known companies in the gaming and social space, such as IGG, Funplus, Cheetah Mobile, Elex, Toutiao, and Momo.
David Ting said, “DataVisor’s results show a detailed breakdown of the app installation channels, and our customers can make more reasonable and sensible decisions by comparing different media and channels, saving millions of dollars per year in fraudulent app installs. DataVisor not only helps advertisers prevent financial losses and improve return on investment, but also ensures the healthy growth of their user base. In our partnership with one of the world’s top mobile app developers, DataVisor found that its mobile marketing budget suffered heavily from fraudulent app installs. The tools used at that time only detected 10% of fraudulent behaviors and could not effectively stop the fraud crisis. After DataVisor’s unsupervised machine learning solution was applied, the company was able to detect 300,000-400,000 fraudulent app installs per month.”
About David Ting
David Ting, Vice President of DataVisor Technology, has over 20 years of leadership experience in mobile development, scalable architecture, cloud infrastructure, big data, and machine learning. He is dedicated to the availability and commercialization of cutting-edge innovations. David has served as the Head of Online Production for Blizzard Entertainment, the General Manager of IGN Entertainment, and the General Manager of NetEase North America. During his time at NetEase, David produced five games, three of which were recommended by Google/Apple. David holds 6 patents and has received 2 IBM Outstanding Achievement Awards and the AltaVista Annual Staff Award.
DataVisor is the leading fraud detection solution designed to uncover modern organized fraud attacks using DataVisor’s proprietary unsupervised machine learning (UML) approach. DataVisor excels at finding unknown fraud attacks by looking for suspicious correlation across all accounts without a need for prior fraud loss labels, and often before they can do damage. The DataVisor Platform provides advanced fraud management with accurate detection score, advanced fraud insights console and custom reporting, as well as flexible deployment and integration options. To date, DataVisor has processed over 800 billion events, detected 218M bad accounts, and protected over 4 billion user accounts from some of the largest financial institutions and online services in the world, such as Yelp, Pinterest, Alibaba and more.