Machine learning and artificial intelligence boosting automotive threat intelligence
- In this work we describe an approach to Threat Intelligence activities for Automotive Products. Taking for granted the increased need to have efficient and actionable results of threat and vulnerability intelligence for automotive products (e.g., because of recent regulations UNECE (United Nations Economic Commission for Europe) R155 and norms ISO/SAE 21434), we successfully applied Natural Language modelling and data science techniques to virtual forums devoted to automotive products. This supplies a concrete way to respond in a fast, automated, maintainable, scalable, and cost-effective way to the questions arising for OEMs and suppliers in the area of threat and vulnerabilities intelligence. Examples of questions are: which is my most hacked product? Which functionalities of my products are mostly hacked? The idea and results presented in this work show the path for the next steps needed to implement a complete framework for automotive threat intelligence based on artificial intelligence.
| Author: | Luca BertoglioGND, Valentina PensoGND, Cosimo Senni Guidotti MagnaniGND |
|---|---|
| Parent Title (English): | 20th escar Europe - The World's Leading Automotive Cyber Security Conference (15. - 16.11.2022) |
| Document Type: | Part of a Book |
| Language: | English |
| Date of Publication (online): | 2022/10/21 |
| Date of first Publication: | 2022/10/21 |
| Publishing Institution: | Ruhr-Universität Bochum, Universitätsbibliothek |
| Tag: | Automotive Cyber Security; Automotive Forums; ISO/SAE 21434; NLTP; R155; Threat Intelligence; Word2vec; vulnerability monitoring |
| First Page: | 39 |
| Last Page: | 49 |
| Dewey Decimal Classification: | Allgemeines, Informatik, Informationswissenschaft / Informatik |
| Konferenz-/Sammelbände: | 20th escar Europe - The World's Leading Automotive Cyber Security Conference |
