A LLM - Based Agent Workflow for Financial Decision Support Using Financial Databases

Kejie Wang

doi:10.54691/7z458936

Authors

Kejie Wang

DOI:

https://doi.org/10.54691/7z458936

Keywords:

Master of law, financial decision-making, financial database.

Abstract

The rapid growth and increasing complexity of financial databases have made it challenging for users to efficiently retrieve and interpret relevant information for decision-making, particularly when queries are expressed in natural language. Traditional financial analysis processes often rely on human experts to translate user intentions into structured data queries, which introduces subjectivity, limited scalability, and operational inefficiencies. Existing automated financial systems, on the other hand, lack sufficient semantic understanding to effectively bridge natural language input and structured financial databases. This paper proposes an LLM-based agent workflow for financial decision support using financial databases. The proposed framework employs large language models to interpret unstructured user queries and transform them into structured prompts aligned with predefined financial data categories. By organizing financial databases into a hierarchical dictionary of semantic labels, the workflow enables accurate data mapping, automated retrieval, and integrated analytical reasoning within a unified system architecture. The agent-based design allows different functional components to collaborate sequentially, supporting flexible and scalable financial information processing. The proposed workflow is presented as an applied conceptual framework, emphasizing system design and workflow integration rather than algorithmic optimization. An illustrative example is provided to demonstrate the workflow execution and evidence-grounded output generation.

Downloads

Download data is not yet available.

References

[1] H. Chen, R. H. L. Chiang, and V. C. Storey, “Business intelligence and analytics: From big data to big impact,” MIS Quarterly, vol. 36, no. 4, pp. 1165–1188, 2012.

[2] D. Arnott and G. Pervan, “A critical analysis of decision support systems research,” Journal of Information Technology, vol. 29, no. 4, pp. 293–306, 2014.

[3] J. Li, Z. Liu, S. Wang, and J. Zhang, “Natural language interfaces for data analytics: A survey,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 4, pp. 3571–3586, 2023.

[4] D. J. Power, Decision Support Systems: Concepts and Resources for Managers. Westport, CT, USA: Quorum Books, 2002.

[5] X. Yang, Y. Liu, and H. Zhang, “AI-enabled decision support systems in financial services: A review,” Decision Support Systems, vol. 153, p. 113668, 2022.

[6] Y. Zhang, X. Yang, and X. Chen, “Large language models in finance: A survey,” ACM Computing Surveys, vol. 55, no. 12, Art. no. 254, 2023.

[7] T. B. Brown, B. Mann, N. Ryder, et al., “Language models are few-shot learners,” in Advances in Neural Information Processing Systems, vol. 33, pp. 1877–1901, 2020.

[8] F. Doshi-Velez and B. Kim, “Towards a rigorous science of interpretable machine learning,” arXiv:1702.08608, 2017.

[9] N. R. Jennings, “On agent-based software engineering,” Artificial Intelligence, vol. 117, no. 2, pp. 277–296, 2000.

[10] X. Xu, X. Wang, Y. Li, and M. Haghighi, “Intelligent decision support systems: A survey,” Journal of Intelligent Information Systems, vol. 52, pp. 267–293, 2019.

[11] D. Araci, “FinBERT: Financial sentiment analysis using pre-trained language models,” arXiv:1908.10063, 2019.

[12] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. NAACL-HLT, Minneapolis, MN, USA, 2019, pp. 4171–4186.

[13] S. Alter, “Nothing is more practical than a good theory: A work system theory perspective,” Journal of the Association for Information Systems, vol. 18, no. 9, pp. 671–690, 2017.

[14] M. Wooldridge, An Introduction to Multiagent Systems, 2nd ed. Chichester, U.K.: Wiley, 2009.

[15] V. Dignum, Responsible Artificial Intelligence: How to Develop and Use AI in a Responsible Way. Cham, Switzerland: Springer, 2019.

[16] R. Guidotti, A. Monreale, S. Ruggieri, F. Turini, F. Giannotti, and D. Pedreschi, “A survey of methods for explaining black box models,” ACM Computing Surveys, vol. 51, no. 5, Art. no. 93, 2018.

[17] Y. R. Shrestha, S. M. Ben-Menahem, and G. von Krogh, “Organizational decision-making with AI,” Journal of Management Information Systems, vol. 36, no. 3, pp. 729–765, 2019.

[18] H. J. Watson, “Tutorial: Business intelligence and analytics,” Communications of the Association for Information Systems, vol. 34, pp. 1245–1248, 2014.