create signature rule with the results of machine learning methods
blend ajazi via Snort-sigs <[email protected]>
| Newsgroups | gmane.comp.security.ids.snort.sigs |
|---|---|
| Message-ID | <DU0P195MB147384162F9C7E53D26BA3C1D7659@DU0P195MB1473.EURP195.PROD.OUTLOOK.COM> |
Hello Snort Community, I am a student who just started working with snort. Below is a result of a decision tree (DT): (The target of the DT is to determine whether a dataset is a cyber-attack or not.) 1 if (N_IN_Conn_P_DstIP > 99.5) and (N_IN_Conn_P_SrcIP <= 99.5) and (max > 0.213) and (srate <= 4.86) and (srate > 0.024) and (max > 2.163) and (srate <= 1.112) and (max <= 4.109) and (max <= 4.095) and (srate <= 0.805) then class: DDoS (proba: 100.0%) | based on 102,730 samples Feature Selection Description N_IN_Conn_P_SrcIP Number of inbound connections per source IP. N_IN_Conn_P_DstIP Number of inbound connections per destination IP. Srate Source-to-destination packets per second Max Maximum duration of aggregated records How can I convert this DT-output in a signature rule in snort? drop tcp any any -> any any ( msg: "DDoS attack" if (N_IN_Conn_P_DstIP > 99.5) and (N_IN_Conn_P_SrcIP <= 99.5) and (max > 0.213) and (srate <= 4.86) and (srate > 0.024) and (max > 2.163) and (srate <= 1.112) and (max <= 4.109) and (max <= 4.095) and (srate <= 0.805) sid:100001 ) I would appreciate any direction of approach with this task. Thank you for the support! Mit freundlichen Grüßen Best regards Blend Ajazi _______________________________________________ Snort-sigs mailing list [email protected] https://lists.snort.org/mailman/listinfo/snort-sigs Please visit http://blog.snort.org for the latest news about Snort! Please follow these rules: https://snort.org/faq/what-is-the-mailing-list-etiquette Visit the Snort.org to subscribe to the official Snort ruleset, make sure to stay up to date to catch the most <a href=" https://snort.org/downloads/#rule-downloads">emerging threats</a>!