Better Call Graphs: A New Dataset of Function Call Graphs for Malware Classification

ICLR 2025 Conference Submission12653 Authors

Keywords: Malware classification, FCG

TL;DR: New Dataset for Malware Classification and Graph Classification

Abstract:

Malware classification by using function call graphs (FCG) is an important task in cybersecurity.
One big challenge in this direction is the lack of representative, large, and unique FCG datasets.
Existing datasets typically contain obsolete Android application packages (APKs), largely consist of small graphs, and include many duplicate FCGs due to repackaging.
This results in misleading graph classification performance.
In this paper, we propose a new comprehensive dataset, Better Call Graphs (BCG), that contains large and unique FCGs from recent APKs, along with graph-level APK features, with benign and malware samples from different types and families.
We establish the necessity of BCG through the evaluation of several baseline approaches on existing datasets.
BCG is available at https://iclr.me.

Primary Area: datasets and benchmarks

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 12653