Individual Packet Features are a Risk to Model Generalisation in ML-Based Intrusion Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908847877128192 |
|---|---|
| author | Kostas, Kahraman Just, Mike Lones, Michael A. |
| author_facet | Kostas, Kahraman Just, Mike Lones, Michael A. |
| contents | Machine learning is increasingly used for intrusion detection in IoT networks. This paper explores the effectiveness of using individual packet features (IPF), which are attributes extracted from a single network packet, such as timing, size, and source-destination information. Through literature review and experiments, we identify the limitations of IPF, showing they can produce misleadingly high detection rates. Our findings emphasize the need for approaches that consider packet interactions for robust intrusion detection. Additionally, we demonstrate that models based on IPF often fail to generalize across datasets, compromising their reliability in diverse IoT environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_07578 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Individual Packet Features are a Risk to Model Generalisation in ML-Based Intrusion Detection Kostas, Kahraman Just, Mike Lones, Michael A. Cryptography and Security Artificial Intelligence Networking and Internet Architecture Machine learning is increasingly used for intrusion detection in IoT networks. This paper explores the effectiveness of using individual packet features (IPF), which are attributes extracted from a single network packet, such as timing, size, and source-destination information. Through literature review and experiments, we identify the limitations of IPF, showing they can produce misleadingly high detection rates. Our findings emphasize the need for approaches that consider packet interactions for robust intrusion detection. Additionally, we demonstrate that models based on IPF often fail to generalize across datasets, compromising their reliability in diverse IoT environments. |
| title | Individual Packet Features are a Risk to Model Generalisation in ML-Based Intrusion Detection |
| topic | Cryptography and Security Artificial Intelligence Networking and Internet Architecture |
| url | https://arxiv.org/abs/2406.07578 |