SIADS: Innovation-Based Anomaly Detection in Streaming Communication Graphs

Research Presentation of Mr. Tanish Verma

Mr. Tanish Verma

Mr. Tanish Verma

M.Tech

Department of Computer Science and Engineering

Paper: SIADS: Innovation-Based Anomaly Detection in Streaming Communication Graphs

Conference: SPCOM 2026

Venue: Indian Institute of Science (IISc), Bengaluru

Research Presentation at SPCOM 2026

Mr. Tanish Verma, M.Tech, from the Department of Computer Science and Engineering, presented his paper “SIADS: Innovation-Based Anomaly Detection in Streaming Communication Graphs” in the poster session at SPCOM 2026, held at Indian Institute of Science (IISc) , Bengaluru.

The paper studies anomaly detection in streaming communication graphs, where users form interaction patterns over time and new interactions need to be evaluated as normal or anomalous.

Research Methodology

SIADS uses an innovation-based scoring approach to measure the change introduced by each incoming edge and updates node representations incrementally instead of retraining the model from scratch.

Presentation Highlights

The presentation provided an opportunity to discuss the work with researchers in graph learning, anomaly detection, and real-time AI systems.

SPCOM 2026 Highlights

Tanish Verma - SPCOM 2026