Tracing Content Requirements in Financial Documents using Multi-granularity Text Analysis

TL;DR

This work introduces FITI, a multi-granularity text-analysis framework for tracing content requirements in financial documents such as UCITS prospectuses. FITI combines IR-based similarity and ML-based feature modeling to identify sentences that express specific information types, using a four-step pipeline: preprocessing, candidate-sentence selection, fine-grained analysis, and a heuristic sentence selector guided by domain phrases. In experiments on 100 real prospectuses and five information types, FITI achieves an average precision of 0.824, recall of 0.646, and F1 of 0.716, significantly outperforming keyword and transformer baselines, and it can detect about 78–80% of missing information types for regulatory compliance tasks. The approach demonstrates the value of integrating similarity- and feature-based analyses with domain knowledge to support automated regulatory checks, while also discussing limitations and opportunities for expanding to other regulatory documents and user studies.

Abstract

The completeness (in terms of content) of financial documents is a fundamental requirement for investment funds. To ensure completeness, financial regulators have to spend significant time carefully checking every financial document based on relevant content requirements, which prescribe the information types to be included in financial documents (e.g., the description of shares' issue conditions and procedures). However, existing techniques provide limited support to help regulators automatically identify the text chunks related to financial information types, due to the complexity of financial documents. In this paper, we propose FITI to trace content requirements in financial documents with multi-granularity text analysis. Given a new financial document, FITI first selects a set of candidate sentences for efficient information type identification. Then, to rank candidate sentences, FITI uses a combination of rule-based and data-centric approaches, by leveraging information retrieval (IR) and machine learning (ML) techniques that analyze the words, sentences, and contexts related to an information type. Finally, a heuristic-based selector, which considers both the sentence ranking and domain-specific phrases, determines a list of sentences corresponding to each information type. We evaluated FITI by assessing its effectiveness in tracing financial content requirements in 100 real-world financial documents. Experimental results show that FITI is able to provide accurate identification with average precision, recall, and F1-score values of 0.824, 0.646, and 0.716, respectively. The overall accuracy of FITI significantly outperforms the best baseline (based on a transformer language model) by 0.266 in terms of F1-score. Furthermore, FITI can help regulators detect about 80% of missing information types in financial documents

Figures (5)

• Figure 1: A Snippet of Financial Document
• Figure 2: Workflow of pre-processing and candidate sentence identification
• Figure 3: Workflow of fine-grained sentence analysis
• Figure 4: Impact of similarity-based and feature-based analyses for FITI
• Figure 5: Impact of the size of the training set for FITI