Android app for scanning your device for potentially harmful applications based on the mal_ai predictive machine learning models.
The Android app is implemented in the Ionic framework using Vue.js and Java. The UI and all UI functions (e.g. click events. Data upload) are implemented using Typescript and HTML5. All functionalities related to finding the apps are implemented in Java. After opening the app, the smartphone is immediately searched for apps, whereby apps that are included by default in Android (e.g. the phone app) are filtered out. The apps found are then listed in the UI, but for data protection reasons nothing is uploaded to the APK download server yet. When the user clicks on "Check apps" the apps are sent to the mal2_ai machine learning predictive backend for in depth analysis.
The MAL2 project applies Deep Neural Networks and Unsupervised Machine Learning to advance cybercrime prevention by a) automating the discovery of fraudulent eCommerce and b) detecting Potentially Harmful Apps (PHAs) in Android. The goal of the MAL2 project is to provide (i) an Open Source framework and expert tools with integrated functionality along the required pipeline – from malicious data archiving, feature selection and extraction, training of Machine Learning classification and detection models towards explainability in the analysis of results (ii) to execute its components at scale and (iii) to publish an annotated Ground-Truth dataset in both application domains. To raise awareness for cybercrime prevention in the general public, two demonstrators, a Fake-Shop Detection Browser Plugin as well as a Android Malware Detection Android app are released that allow live-inspection and AI based predictions on the trustworthiness of eCommerce sites and Android apps.
The work is based on results carried out in the research project MAL2 project, which was partially funded by the Austrian Federal Ministry for Climate Action, Environment, Energy, Mobility, Innovation and Technology (BMK) through the ICT of the future research program (6th call) managed by the Austrian federal funding agency (FFG).
- Austrian Institute of Technology GmbH, Center for Digital Safety and Security AIT
- Austrian Institute for Applied Telecommunications ÖIAT
- X-NET Services GmbH XNET
- Kuratorium sicheres Österreich KSÖ
- IKARUS Security Software IKARUS
More information is available at www.malzwei.at
For details on behalf of the MAL2 consortium contact: Andrew Lindley (project lead) Research Engineer, Data Science & Artificial Intelligence Center for Digital Safety and Security, AIT Austrian Institute of Technology GmbH Giefinggasse 4 | 1210 Vienna | Austria T +43 50550-4272 | M +43 664 8157848 | F +43 50550-4150 andrew.lindley@ait.ac.at | www.ait.ac.at or Woflgang Eibner, X-NET Services GmbH, we@x-net.at
The MAL2 Software stack is dual-licensed under commercial and open source licenses. The Software in this repository is subject of the terms and conditions defined in file 'LICENSE.md'