{"paper":{"title":"ML Defender (aRGus NDR): An Open-Source Embedded ML NIDS for Botnet and Anomalous Traffic Detection in Resource-Constrained Organizations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"An open-source ML network detector on low-cost hardware detects botnet traffic with F1 of 0.9985 where signature and scripted systems largely fail.","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Alonso Isidoro Rom\\'an","submitted_at":"2026-04-03T05:20:13Z","abstract_excerpt":"Ransomware and DDoS attacks disproportionately impact hospitals, schools, and small organizations that cannot afford enterprise security. We present ML Defender (aRGus NDR), an open-source C++20 NIDS with embedded ML inference, deployable on commodity hardware at 150-200 USD.\n  The system implements a six-component pipeline over eBPF/XDP, ZeroMQ, and Protocol Buffers, with a dual-score Fast Detector + Random Forest architecture. Evaluated on CTU-13 Neris: F1=0.9985, Precision=0.9969, Recall=1.0000 (2 FP in 12,075 benign flows, both VirtualBox artifacts).\n  We report the first three-paradigm ex"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"aRGus NDR achieves F1=0.9985, Precision=0.9969, Recall=1.0000 on CTU-13 Neris (2 FP in 12,075 benign flows) while Suricata 6.0.10 with 50,010 rules generates zero alerts and Zeek 8.1.2 achieves F1=0.042.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The CTU-13 Neris dataset and the offline confirmation experiment on 323,154 packets constitute a fair, representative, and artifact-free test of real-world botnet detection performance across the three paradigms.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ML Defender achieves F1=0.9985 on CTU-13 Neris botnet detection with a dual fast-detector plus random forest model, outperforming Suricata (zero alerts) and Zeek (F1=0.042) in a three-paradigm comparison.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"An open-source ML network detector on low-cost hardware detects botnet traffic with F1 of 0.9985 where signature and scripted systems largely fail.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"31ee405c26a9c5cef975c669f82a4f5bcb57c2185f1e264c06a55ea8c07cffff"},"source":{"id":"2604.04952","kind":"arxiv","version":6},"verdict":{"id":"8f2d4423-3b44-49b8-80c2-3175ce760be5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T20:21:18.635924Z","strongest_claim":"aRGus NDR achieves F1=0.9985, Precision=0.9969, Recall=1.0000 on CTU-13 Neris (2 FP in 12,075 benign flows) while Suricata 6.0.10 with 50,010 rules generates zero alerts and Zeek 8.1.2 achieves F1=0.042.","one_line_summary":"ML Defender achieves F1=0.9985 on CTU-13 Neris botnet detection with a dual fast-detector plus random forest model, outperforming Suricata (zero alerts) and Zeek (F1=0.042) in a three-paradigm comparison.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The CTU-13 Neris dataset and the offline confirmation experiment on 323,154 packets constitute a fair, representative, and artifact-free test of real-world botnet detection performance across the three paradigms.","pith_extraction_headline":"An open-source ML network detector on low-cost hardware detects botnet traffic with F1 of 0.9985 where signature and scripted systems largely fail."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.04952/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}