Financial reports such as 10-K reports contain extensive information about a company’s financial performance, risks, and operations. However, these documents are often lengthy and complex, making manual analysis difficult and time-consuming for investors and analysts. Traditional methods struggle to efficiently extract meaningful insights from unstructured financial text, leading to challenges in effective decision-making. To address these limitations, Finenzo is proposed as a system that utilizes Natural Language Processing (NLP) techniques to analyze financial statements and extract useful insights. The system retrieves 10-K reports from the SEC EDGAR database, performs text preprocessing, and applies TF-IDF for feature extraction. A financial sentiment dictionary is used to identify sentiment patterns and generate quantitative scores from textual data. These insights are combined with financial indicators to support better evaluation of company performance and assist in data-driven decision-making. By converting unstructured financial data into structured information, the system improves analysis efficiency and provides a scalable approach for financial analysis.
Chavan et al. (2026) studied this question.