Curated list of the leading open‑source fraud detection platforms that support rule‑based customization for real‑time and batch analysis.
Get targeted exposure with custom position pinning and highlighted placement.
Scalable threat‑detection platform built on Hadoop and Spark that lets you define custom detection rules for network‑traffic anomalies and fraud patterns.
Real‑time security analytics framework that integrates a pluggable rule engine (Storm + Elasticsearch) for detecting fraudulent activities across streaming data.
Powerful Business Rule Management System (BRMS) written in Java; widely used to encode complex fraud‑detection logic that can be changed without redeploying code.
High‑performance Complex Event Processing (CEP) engine that enables rule‑based detection of suspicious event patterns in financial transactions.
Stream‑processing framework with built‑in CEP library; supports custom fraud‑detection rules that run on low‑latency data pipelines.
Lightweight stream‑processing library and SQL‑like interface for defining fraud‑detection rules directly on Kafka topics.
Open‑source rule engine that lets business users write fraud‑detection rules in Excel‑like tables, which are compiled to Java at runtime.
GitHub project that combines scikit‑learn models with a rule‑based pre‑filter; fully configurable via YAML rule files.
Python toolbox for scalable outlier detection; can be wrapped with custom rule logic to flag fraudulent records.
Distributed machine‑learning platform that includes auto‑ML pipelines; rules can be added post‑model to enforce business constraints.
Search and analytics engine with a rule‑based alerting feature (Watcher) that can trigger on anomalous transaction patterns.
Open‑source CEP engine from WSO2; provides a SQL‑like language for defining fraud‑detection rules over streaming data.
Community‑driven Python library focused on e‑commerce fraud detection; supports rule‑based scoring and model integration.
R package that implements statistical and rule‑based techniques for credit‑card fraud detection, with easy configuration via JSON rules.
Hierarchical Temporal Memory implementation for anomaly detection; can be combined with custom rule layers to flag fraud.
Kubernetes‑native platform for deploying ML models; includes a rule‑engine sidecar that can enforce business rules on model predictions.
Open‑source deep‑learning server that exposes models via REST; supports rule‑based post‑processing to filter fraudulent scores.
While focused on threat intel, MISP’s correlation engine can be repurposed with custom rules to detect fraud‑related indicators.
Java‑based rule mining library that can generate interpretable fraud‑detection rules from historical data and apply them in production.
Open‑source anomaly detection framework (Python) that provides rule‑based thresholds for real‑time fraud monitoring.