Paper accepted at MSR 2026: AFGNN – API Misuse Detection using Graph Neural Networks and Clustering

Congratulations to the authors Ponnampalam Pirapuraj, Tamal Mondal, Sharanya Gupta, Akash Lal, Somak Aditya, and Dr.Jyothi Vedurada on this remarkable achievement!

We are pleased to announce that the paper titled “AFGNN: API Misuse Detection using Graph Neural Networks and Clustering” has been accepted at the 23rd International Conference on Mining Software Repositories (MSR 2026), co-located with ICSE 2026 in Rio de Janeiro, Brazil.

Key Highlight

This work introduces AFGNN, a graph neural network–based approach that combines a novel API Flow Graph representation with clustering to detect real-world API misuses. The method achieves significant improvements over state-of-the-art techniques, advancing the reliability of software development tools.

Congratulations to all the authors for this impactful research contribution!