Research

The beta version of the Fake Shop Detector is the result of ongoing research projects. Work on the detection of fake shops by artificial intelligence has been ongoing since 2018. The data set on fraudulent online shops used for the development of the AI models is available for research purposes upon request.

The aim of the SINBAD project is to gain new insights into proactive fake shop detection through innovative, user-centred survey methods, data-supported models and the deepening of machine learning procedures and to develop effective counter-narratives with the involvement of the BMSGPK, the body responsible for the needs, in order to empower consumers. The project is funded by the Federal Ministry of Agriculture, Forestry, Environment and Water Management in the KIRAS funding programme 2020.

Project name: SINBAD - Security and prevention of organised Internet order fraud for users through digital forensics measures

Website: https://projekte.ffg.at/projekt/3807747

Consortium:Austrian  Institute for Applied  Telecommunications, Center for Digital Safety & Security, Center for Technology Experience (AIT Austrian Institute of Technology GmbH), X-Net Services GmbH,Ciuvo GmbH

Duration: 2020-2022

The aim of the project is to develop gamification elements for the Detector plug-in. In this way, the Fake Shop Detector is to become a community tool that supports the detection of Internet fraud. Interested users will become valuable agents in the prevention and detection of crime in online commerce through their engagement and use of the tool. Relevant authorities and stakeholders receive early warnings and can delegate tasks to the community. The project is supported by netidee, a funding action of the Internet Foundation Austria.

Project name: DETECT - real-time protection against fake shops through community-driven AI

Website: https://www.netidee.at/detect

Consortium: Center for Digital Safety & Security of the AIT Austrian Institute of Technology GmbH (project management), Austrian Institute for Applied Telecommunications

Duration: 2021

The goal of the INSPECTION project is to automatically identify hacked websites by searching the German-language internet and classifying the sites with artificial intelligence methods from the outside. Methods are also to be developed to inform those affected, to repair the damage and to reduce the risk of future successful attacks. Both those affected and website operators are to be made aware of the problems. The project is funded by the German Federal Ministry of Education and Research as part of KMU Innovativ, and Watchlist Internet is involved as an associated partner.

Project summary: https://www.mindup.de/data-scientists/anwendungsfaelle/fake-online-shops

Consortium: mindUp Web + Intelligence GmbH (project management), Karlsruhe Institute of Technology (KIT), BDO AG, Hamburg, Host Europe GmbH, Allianz für Sicherheit in der Wirtschaft, Deutschland sicher im Netz e.V., ECO Verband - SIWECOS, Baden-Württembergischer Handwerkskammertag, Fachverband Elektro- und Informationstechnik BW, Österreichisches Institut für angewandte Telekommunikation, SWISS INTERNET SECURITY ALLIANCE

Duration: 2020-2023

The goal of MAL2 (MAchine Learning detection of MALicious content) was to use Deep Neural Networks and Unsupervised Learning for the automated detection of fraudulent fake shops and harmful Android Apps (PHAs) and thus to contribute to the improvement of cybercrime prevention. The MAL2 project was supported by funding from the BMVIT in the 6th call for proposals of the Austrian Research Promotion Agency (FFG).

Project website: https://www.malzwei.at/

Consortium: Center forDigital Safety & Security, AIT Austrian Institute of Technology GmbH (project management), Austrian Institute for Applied Telecommunications, X-Net Services GmbH, Kompetenzzentrum Sicheres Österreich, Ikarus Security Software GmbH

Duration: 2018-2020

The aim of the project KOSOH (Consumer Protection in Online Commerce) was to capture the knowledge of Watchlist Internet in the manual detection of fake shops in order to increasingly automate machine learning with neural networks for the automatic classification of websites based on the similarity of their source code features as well as through procedures for the automatic identification of seals of approval through image analysis procedures. The project was funded with financial support from the BMLRT in the KIRAS 2018 funding line.

Project summary: https://www.kiras.at/gefoerderte-projekte/detail/d/kosoh-1

Consortium: Center forDigital Safety & Security, AIT Austrian Institute of Technology GmbH (project management), Austrian Institute for Applied Telecommunications

Duration: 2018-2020