DAgent: A Relational Database-Driven Data Analysis Report Generation Agent
Wenyi Xu, Yuren Mao, Xiaolu Zhang, Chao Zhang, Xuemei Dong, Mengfei Zhang, Yunjun Gao
TL;DR
DAgent addresses relational database‑driven analytical report generation (RDB‑DA) by introducing an LLM‑based agent with planning, tools, and memory to decompose questions, retrieve cross‑table data, and generate concise, complete reports. It furthermore contributes DA‑Dataset, a multi‑domain benchmark (DA‑CCKS and DA‑BIRD) designed to stress cross‑table reasoning and report synthesis, along with evaluation metrics for retrieval and report quality. The system leverages LoRA‑based domain adaptation, combines direct encoding retrieval with Text‑to‑SQL retrieval, and uses a modular pipeline (planning, tools, memory) to optimize end‑to‑end report generation, formalized as $R = \mathrm{LLM}(\text{Prompt})$ and supported by SQL rewriting $SQL_i'$. Experimental results show that DAgent outperforms state‑of‑the‑art baselines in both retrieval accuracy and report quality, highlighting its potential for practical automated database analysis in finance, healthcare, and related domains.
Abstract
Relational database-driven data analysis (RDB-DA) report generation, which aims to generate data analysis reports after querying relational databases, has been widely applied in fields such as finance and healthcare. Typically, these tasks are manually completed by data scientists, making the process very labor-intensive and showing a clear need for automation. Although existing methods (e.g., Table QA or Text-to-SQL) have been proposed to reduce human dependency, they cannot handle complex analytical tasks that require multi-step reasoning, cross-table associations, and synthesizing insights into reports. Moreover, there is no dataset available for developing automatic RDB-DA report generation. To fill this gap, this paper proposes an LLM agent system for RDB-DA report generation tasks, dubbed DAgent; moreover, we construct a benchmark for automatic data analysis report generation, which includes a new dataset DA-Dataset and evaluation metrics. DAgent integrates planning, tools, and memory modules to decompose natural language questions into logically independent sub-queries, accurately retrieve key information from relational databases, and generate analytical reports that meet the requirements of completeness, correctness, and conciseness through multi-step reasoning and effective data integration. Experimental analysis on the DA-Dataset demonstrates that DAgent's superiority in retrieval performance and analysis report generation quality, showcasing its strong potential for tackling complex database analysis report generation tasks.