During the last decade mobile phones and tablets evolved into smart devices with enormous computing power and storage capacity packed in a pocket size. People around the globe have quickly moved from laptops to smartphones for their daily computational needs. From web browsing, social networking, photography to critical bank payments and intellectual property every thing has got into smartphones; and undoubtedly Android has dominated the smartphone market. Android growth also attracted cyber criminals to focus on creating attacks and malwares to target Android users. Malwares in different category are seen in the Android ecosystem, including botnets, Ransomware, click Trojan, SMS frauds, banking Trojans.
Due to huge amount of application being developed and distributed every day, Android needs malware analysis techniques that are different than any other operating system. This research focuses on defining a process of finding Android malware in a given large number of new applications. Research utilizes machine learning techniques in predicting possible malware and further provide assistance in reverse engineering of malware. Under this thesis an assistive Android malware analysis system “AndroSandX” is proposed, researched and developed. AndroSandX allows researcher to quickly analyze potential Android malware and help perform manual analysis.
Key features of the system are strong assistive capabilities using machine learning, built in ticketing system, highly modular design, storage with non-relational databases, backup of analysis data for archival, assistance in manual analysis and threat intelligence. Research results shows that the system has a prediction accuracy of around 92%. Research has wide scope and lean towards providing industry oriented Android malware analysis assistive system/product.
Computing Security (MS)
Department, Program, or Center
Department of Computing Security (GCCIS)
Jadhav, Suyash, "Advance Android PHAs/Malware Detection Techniques by Utilizing Signature Data, Behavioral Patterns and Machine Learning" (2016). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus