In recent years, with the rapid development of mobile Internet technology and the rise and popularization of intelligent mobile devices, the mobile advertising industry has developed rapidly. According to the emarketer report, the total global mobile advertising expenditure has reached 240.95 billion US dollars in 2019, and is expected to continue to grow to 286.5 billion US dollars in 2020. However, with the continuous growth of mobile advertising budget, mobile application installation cheating is becoming increasingly rampant, which not only causes advertising budget waste to advertisers, but also has a serious adverse impact on the advertising platform delivery effect and reputation.
Among all kinds of cheating types of mobile application installation, BOT installation cheating is undoubtedly the most difficult to identify, and it is also a relatively large type of cheating. According to a report of appsflyer in 2019, appsflyer detected more than 1.6 billion installation cheating in the three years from 2017 to 2019, including more than 900 million bots installation cheating.
Bots installation cheating refers to that cheaters use bots to simulate the application installation behavior of real users, so as to steal the advertising budget of advertisers. The reason why bots installation cheating is difficult to detect is that bots looks like a real user. It has real IP, device, etc., and can even carry out events within the application, such as opening the application at a specified time, and purchasing more.
At present, many mobile advertising fraud detection methods based on machine learning are aimed at solving the problem of advertising display or click cheating, but there is little research on the installation of cheating detection, especially the installation of bots cheating detection. According to Dr. Zhu Yadong, vice president of mobvista group, many of the current methods adopt integration methods and other technologies, such as randomforest and xgboost. Although these methods can mine rich cheating patterns through complex feature engineering, they cannot use structural information to build various relationships between entities, so these methods are not the best solution u3002
In this context, mobvista anti cheating team proposed a hybrid learning method, which combines graph neural network (GNN) and gradient promotion classifier. It can consider local context information and global context information at the same time, so as to better detect BOT installation cheating in mobile advertising. Instead of directly using the existing GNN model, the team constructed a heterogeneous graph according to the specific business of mobvista and the characteristics of BOT installation cheating, and designed a novel message passing mechanism to extract local context information; and a pre trained gradient promotion classifier model was used to extract global context information. The architecture of the botspot model is shown in the following figure:
The results show that botvista has more advantages in the detection of the data of botbotspot by installing the model. When the precision is 90%, the recall rate of botspot model is improved by at least 2.2% and 5.75% respectively compared with all other baseline methods on two offline datasets.
As the worlds leading technology platform, mobvista has more than 1 million applications installed every day. Therefore, mobvista attaches great importance to the research on cheating methods of various mobile application installation, so as to prevent advertisers budget from being wasted without any actual conversion. It is worth mentioning that in May this year, mobvista also released white paper 2.0 on anti cheating in mobile advertising aiming at the problem of cheating in mobile advertising, which describes in detail the current situation of cheating in the field of mobile advertising, the methods of cheating and the corresponding anti cheating strategies.
Dr. Zhu Yadong said that the key to enhance the transparency of the mobile advertising industry and promote the standardized development of the industry is to study the cheating detection of bots installation in the field of mobile advertising. However, few institutions have conducted research in this field. Mobvista has always been at the forefront of the field of anti cheating in mobile advertising. It is hoped that through the sharing of this paper, it can serve the industry Bring more thinking.
According to public information, mobvista was founded in 2013 and now covers more than 85 countries around the world, with offices in 18 cities around the world, providing mobile advertising promotion services to more than 3000 application developers worldwide, and more than 45% of its more than 700 employees are technical R & D personnel. The paper that was employed by CIKM was completed under the guidance of Zhu Yadong, vice president of CIKM, with the guidance and help of Professor Liang shangsong of Sun Yat sen University. On October 22, mobvista senior algorithm engineer Yao Tianjun and senior algorithm engineer Li Qing will share the results in detail at CIKM Conference on behalf of the anti cheating team. Please look forward to it. Source: editor in charge of mass news: Chen Tiqiang_ NB6485
According to public information, mobvista was founded in 2013 and now covers more than 85 countries around the world, with offices in 18 cities around the world, providing mobile advertising promotion services to more than 3000 application developers worldwide, and more than 45% of its more than 700 employees are technical R & D personnel. The paper that was employed by CIKM was completed under the guidance of Zhu Yadong, vice president of CIKM, with the guidance and help of Professor Liang shangsong of Sun Yat sen University. On October 22, mobvista senior algorithm engineer Yao Tianjun and senior algorithm engineer Li Qing will share the results in detail at CIKM Conference on behalf of the anti cheating team. Please look forward to it.