Traffic Classification and Application Identification in Network Forensics

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F18%3APU130682" target="_blank" >RIV/00216305:26230/18:PU130682 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-3-319-99277-8" target="_blank" >http://dx.doi.org/10.1007/978-3-319-99277-8</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-99277-8" target="_blank" >10.1007/978-3-319-99277-8</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Traffic Classification and Application Identification in Network Forensics

Popis výsledku v původním jazyce

Network traffic classification is an absolute necessity for network monitoring, security analysis, and digital forensics. Without accurate traffic classification, computation demands on analysis of all IP flows are enormous. Classification can also reduce the number of flows that need to be analyzed, prioritize, and order them for an investigator to analyze the most forensically significant first. This paper presents an automatic feature elimination method based on a feature correlation matrix. Furthermore, we compare two algorithms adapted from literature, that offer high accuracy and acceptable performance, and our algorithm -- Enhanced Statistical Protocol Identification (ESPI). Each of these algorithms is used with a subset of features that best suits it. We evaluate these algorithms on their ability to identify application layer protocols and additionally applications themselves. Experiments show that the Random Forest based classifier yields the most promising results, whereas our algorithm provides an interesting tradeoff between higher performance and slightly lower accuracy.

Název v anglickém jazyce

Traffic Classification and Application Identification in Network Forensics

Popis výsledku anglicky

Network traffic classification is an absolute necessity for network monitoring, security analysis, and digital forensics. Without accurate traffic classification, computation demands on analysis of all IP flows are enormous. Classification can also reduce the number of flows that need to be analyzed, prioritize, and order them for an investigator to analyze the most forensically significant first. This paper presents an automatic feature elimination method based on a feature correlation matrix. Furthermore, we compare two algorithms adapted from literature, that offer high accuracy and acceptable performance, and our algorithm -- Enhanced Statistical Protocol Identification (ESPI). Each of these algorithms is used with a subset of features that best suits it. We evaluate these algorithms on their ability to identify application layer protocols and additionally applications themselves. Experiments show that the Random Forest based classifier yields the most promising results, whereas our algorithm provides an interesting tradeoff between higher performance and slightly lower accuracy.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/VI20172020062" target="_blank" >VI20172020062: Integrovaná platforma pro zpracování digitálních dat z bezpečnostních incidentů (TARZAN)</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Fourteenth Annual IFIP WG 11.9 International Conference on Digital Forensics

ISBN

978-3-319-99277-8

ISSN

1868-4238

e-ISSN

1868-422X

Počet stran výsledku

21

Strana od-do

161-181

Název nakladatele

Springer International Publishing

Místo vydání

New Delhi

Místo konání akce

New Delhi

Datum konání akce

3. 1. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000475838900010