Citation :

Santosh Srinivasaiah, M Rajeswari, "Graph-Augmented AI for Hierarchical Reasoning over Enterprise Codebases and Regulatory Text," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 9, pp. 150-163, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I9P113

Abstract

Graph-Augmented AI Systems are on the cusp of a paradigm shift, marking the next significant leap in deep-level reasoning within risk-complex enterprise code and regulatory documents. As part of this activity, this paper proposes an innovative algorithmic result and solution using the concept of Hierarchy Graph Augmented Reasoning Network (HiGAR-Net), an innovative approach goal to aimed at modelling complex-risk semantics through both graph and deep-level reasoning. Through graph augmentation, the proposed approach is able to provide high interpretability and scalability. Experimental results show that HiGAR-Net achieves a high weighted precision of 95%, a weighted recall of 93%, and a weighted F1 of 95% with good class-wise performance in the compliance classification task. Although the proposed framework is intended to be multimodal extensions friendly, the present experimental analysis of the framework concentrates on graph-based representations based on enterprise codebases and regulatory writing.

Keywords

Graph-Augmented AI, Hierarchical reasoning networks, Enterprise code intelligence, Regulatory document analysis, HiGAR-Net, Semantic dependency modeling.

References

[1] Ivar Nelson, and Oscar Andersson, “Enhancing Code Review at Scale with Generative AI and Knowledge Graphs: An Agentic Graphrag Framework for Enterprise Code Review,” 2025.
[Google Scholar]

[2] Ricky Sun et al., “Graph XAI: Graph-Augmented AI with ADEV,” Paestum’24: 26^th Int’l Workshop on DOLAP, Paestum, Italy, pp. 1-8, 2024.
[Google Scholar]

[3] Zulun Zhu et al., “Graph-based Approaches and Functionalities in Retrieval-Augmented Generation: A Comprehensive Survey,” ACM Computing Surveys, vol. 58, no. 10, pp. 1-38, 2026.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Stefan Wagenpfeil, “Multimedia Graph Codes for Fast and Semantic Retrieval-Augmented Generation,” Electronics, vol. 14, no. 2, pp. 1-25, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[5] Jiaxi Xu, Jun Ai, and Tao Shi, “Software Defect Prediction for Specific Defect Types based on Augmented Code Graph Representation,” 2021 8^th International Conference on Dependable Systems and their Applications (DSA), Yinchuan, China, pp. 669-678, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Qinggang Zhang et al., “A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models,” arxiv preprint, pp. 1-26, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[7] Shengjie Ma et al., “Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-Guided Retrieval Augmented Generation,” International Conference on Learning Representations 2025 (ICLR 2025), pp. 1-25, 2025.
[Google Scholar] [Publisher Link]

[8] Bowen Jin et al., “Large Language Models on Graphs: A Comprehensive Survey,” IEEE Transactions on Knowledge and Data Engineering, vol. 36, no. 12, pp. 8622-8642, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Ziwei Chai et al., “GraphLLM: Boosting Graph Reasoning Ability of Large Language Model,” IEEE Transactions on Big Data, vol. 12, no. 2, pp. 475-483, 2026.
[CrossRef] [Google Scholar] [Publisher Link]

[10] Bowen Jin et al., “Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs,” Findings of the Association for Computational Linguistics: ACL, Association for Computational Linguistics, Bangkok, Thailand, pp. 163-184, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Hany F. Atlam, “LLMs in Cyber Security: Bridging Practice and Education,” Big Data and Cognitive Computing, vol. 9, no. 7, pp. 1-53, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[12] Agada Joseph Oche et al., “A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions,” arxiv preprint, pp. 1-33, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Yihan Li et al., “Mitigating Hallucination in Large Language Models (LLMs): An Application-Oriented Survey on RAG, Reasoning, and Agentic Systems,” arxiv preprint, pp. 1-25, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[14] Kaize Ding et al., “Data Augmentation for Deep Graph Learning: A Survey,” ACM SIGKDD Explorations Newsletter, vol. 24, no. 2, pp. 61-77, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[15] Ashish Vaswani et al., “Attention is All You Need,” arxiv preprint, pp. 1-15, 2017.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Petar Veličković et al., “Graph Attention Networks,” arxiv preprint, pp. 1-12, 2017.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Miltiadis Allamanis, Marc Brockschmidt, and Mahmoud Khademi, “Learning to Represent Programs with Graphs,” arxiv preprint, pp. 1-17, 2018.
[CrossRef] [Google Scholar] [Publisher Link]

[18] Rex Ying et al., “Hierarchical Graph Representation Learning with Differentiable Pooling,” Proceedings of the 32^nd International Conference on Neural Information Processing Systems, Curran Associates Inc. 57 Morehouse Lane, Red Hook, NY, United States, pp. 4805-4815, 2018.
[Google Scholar] [Publisher Link]

[19] Alejandro Barredo Arrieta et al., “Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges Toward Responsible AI,” Information Fusion, vol. 58, pp. 82-115, 2020.
[CrossRef] [Google Scholar] [Publisher Link]