Smartphones and mobile tablets are rapidly becoming indispensable in daily life. Android has been the most popular mobile operating system since 2012. However, owing to the open nature of Android, countless malwares are hidden in a large number of benign apps in Android markets that seriously threaten Android security. Deep learning is a new area of machine learning research that has gained increasing attention in artificial intelligence. In this study, we propose to associate the features from the static analysis with features from dynamic analysis of Android apps and characterize malware using deep learning techniques. We implement an online deep-learning-based Android malware detection engine (DroidDetector) that can automatically detect whether an app is a malware or not. With thousands of Android apps, we thoroughly test DroidDetector and perform an in-depth analysis on the features that deep learning essentially exploits to characterize malware. The results show that deep learning is suitable for characterizing Android malware and especially effective with the availability of more training data. DroidDetector can achieve 96.76% detection accuracy, which outperforms traditional machine learning techniques. An evaluation of ten popular anti-virus softwares demonstrates the urgency of advancing our capabilities in Android malware detection.

Traffic classification is critical to effective network management. However, more and more proprietary, encrypted, and dynamic protocols make traditional traffic classification methods less effective. A Message and Command Correlation (MCC) method was developed to identify interactive protocols (such as P2P file sharing protocols and Instant Messaging (IM) protocols) by session analyses. Unlike traditional packet-based classification approaches, this method exploits application session information by clustering packets into application messages which are used for further classification. The efficacy and accuracy of the MCC method was evaluated with real world traffic, including P2P file sharing protocols Thunder and BitTorrent, and IM protocols QQ and GTalk. The tests show that the false positive rate is less than 3% and the false negative rate is below 8%, and that MCC only needs to check 8.7% of the packets or 0.9% of the traffic. Therefore, this approach has great potential for accurately and quickly discovering new types of interactive application protocols.