Idea
A real-time anomaly detection platform for streaming graph edge data that enhances cybersecurity and fraud detection accuracy.
Research Paper
Core Innovation
This paper introduces ADAPTIVE-GRAPHSKETCH, which combines multi-layer tensor sketching with Count-Min Sketch using Conservative Update to efficiently track edge frequencies in streaming graphs. It uniquely integrates Bayesian inference for probabilistic anomaly scoring and adaptive thresholding via EWMA to handle bursty traffic patterns. This approach improves scalability, interpretability, and adaptability compared to prior methods.
Market Size (TAM)
$10–20B TAM, $2–5B SAM; assumption: growing demand for real-time cybersecurity and fraud detection in large-scale networked systems.
Potential Customers & Pain Points
- Cybersecurity firms needing scalable intrusion detection
- Power grid operators monitoring network anomalies
- Financial institutions detecting fraud in transaction graphs
- Network administrators managing large-scale streaming data
- AI developers requiring efficient graph anomaly APIs
Business Model
SaaS platform offering API access for real-time anomaly detection with tiered pricing based on data volume and feature set.
Competitive Landscape
- ANOEDGE-G/L
- MIDAS-R
- F-FADE
Implementation Challenges
- Integration with existing enterprise security infrastructure
- Handling diverse and evolving graph data types
- Convincing customers to switch from established anomaly detection tools
Validation Strategy
- Pilot deployments with cybersecurity firms to benchmark detection accuracy and speed
- Performance testing on diverse real-world streaming datasets
- Customer feedback cycles to refine adaptive thresholding and usability
Research Paper Overview
Adaptive-GraphSketch: Real-Time Edge Anomaly Detection via Multi-Layer Tensor Sketching and Temporal Decay
Summary
Anomaly detection in dynamic graphs is essential for identifying malicious activities, fraud, and unexpected behaviors in real-world systems such as cybersecurity and power grids. However, existing approaches struggle with scalability, probabilistic interpretability, and adaptability to evolving traffic patterns. In this paper, we propose ADAPTIVE-GRAPHSKETCH, a lightweight and scalable framework for real-time anomaly detection in streaming edge data. Our method integrates temporal multi-tensor sketching with Count-Min Sketch using Conservative Update (CMS-CU) to compactly track edge frequency patterns with bounded memory, while mitigating hash collision issues. We incorporate Bayesian inference for probabilistic anomaly scoring and apply Exponentially Weighted Moving Average (EWMA) for adaptive thresholding tuned to burst intensity. Extensive experiments on four real-world intrusion detection datasets demonstrate that ADAPTIVE-GRAPHSKETCH outperforms state-of-the-art baselines such as ANOEDGE-G/L, MIDAS-R, and F-FADE, achieving up to 6.5% AUC gain on CIC-IDS2018 and up to 15.6% on CIC-DDoS2019, while processing 20 million edges in under 3.4 seconds using only 10 hash functions. Our results show that ADAPTIVE-GRAPHSKETCH is practical and effective for fast, accurate anomaly detection in large-scale streaming graphs.