Distress detection (poster abstract)

0Citations
Citations of this article
4Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Web attacks are a major security concern as novel attacks can be easily created by exploiting different vulnerabilities, using different attack payloads, and/or encodings (obfuscation). Intrusion detection systems (IDS) aim to correctly detect attacks. There are two main approaches to intrusion detection: misuse and anomaly detection. Despite the difference in approach, they both fail to offer adequate resilience to novel attacks due to the difficulty in generalizing beyond known attack or normal behavior [1]. © 2012 Springer-Verlag.

Cite

CITATION STYLE

APA

Vella, M., Terzis, S., & Roper, M. (2012). Distress detection (poster abstract). In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7462 LNCS, pp. 384–385). https://doi.org/10.1007/978-3-642-33338-5_24

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free