Statistical Decision Making for Authentication and Intrusion Detection - Statistics > Machine LearningReport as inadecuate




Statistical Decision Making for Authentication and Intrusion Detection - Statistics > Machine Learning - Download this document for free, or read online. Document in PDF available to download.

Abstract: User authentication and intrusion detection differ from standardclassification problems in that while we have data generated from legitimateusers, impostor or intrusion data is scarce or non-existent. We review existingtechniques for dealing with this problem and propose a novel alternative basedon a principled statistical decision-making view point. We examine thetechnique on a toy problem and validate it on complex real-world data from anRFID based access control system. The results indicate that it cansignificantly outperform the classical world model approach. The method couldbe more generally useful in other decision-making scenarios where there is alack of adversary data.



Author: Christos Dimitrakakis, Aikaterini Mitrokotsa

Source: https://arxiv.org/



DOWNLOAD PDF




Related documents