From Understanding to Problem Solving： Applying Deep Learning to Defend Against the Deceptive Advertising and Phone Scams

摘要：

The advance of mobile equipment and network technology boosts the need of mobile marketing and mobile advertising. As the number of installed our featured Apps has reached approximately 3 billions while the number of active users is roughly 700 millions, we found that the deceptive advertising (deceptive ads) with the use of false or misleading statements in advertising texts and figures varies with regions and time zones. The deceptive ads trick users to install unnecessary Apps and will cause the reputation loss of the advertisers. Moreover, phone scams also are another annoying situation. However, the detection of such deceptive ads and phone scams is a challenging task; deceptive ads exhibit fast-flux behavior, while phone scams exhibit caller ID spoofing behavior and therefore is more difficult to be caught.

Alpha GO has proved the value of deep learning in pattern recognition. Based on our customized feature extraction and fine-tune model, will share with you in this talk our experience in developing effective deep learning mechanism for detecting deceptive ads and phone scams. Different with academic research, our proposed system has been deployed in our backend and featured products for intensive analysis and has shown that such hybrid approach yields desired results based on our massive real dataset. Moreover, We will also share with you how to do the right things to increase the research quality with the right ways to do the mobile and internet security.

講者簡介：

Hsien-De Huang (a.k.a. TonTon) is working for Leopard Mobile and a Ph. D. candidate in the Dept. Computer Science and Information Engineering at National Cheng-Kung University, Taiwan. For the past few years, he was a Software Developer at Verint Systems (Taiwan), Senior Security Engineer at Acer e-Enabling Data Center (Acer eDC) and Project Assistant Researcher at the National Center for High-Performance Computing (NCHC).

He received his M.S. degree from the Dept. Information and Learning Technology , National University of Tainan, Taiwan. He also was a visiting Ph. D student in the UK for research project “2010 Initiative Research Cooperation among Top Universities between UK and Taiwan” at University of Essex, UK and in the research project “2012 NSC-INRIA International Program – Associate Team (II)” at INRIA Saclay, France. His current major research interests include Deep Learning, Malware Analysis, Android Reverse-Engineering, Type-2 Fuzzy Logic, and Ontology Applications.
His personal website is http://TWMAN.ORG

• ← High performance hardware accelerator for optimizing Hadoop
• Achieve Big Data Analytic Platform with Lambda Architecture on Cloud →