Explainable Risk Classification in Financial Reports
FinBERT-XRC model uses explainable AI for risk classification in 10-K financial reports, offering multi-level explanations.
What it examines
This paper introduces FinBERT-XRC, an explainable deep-learning model designed to assess post-event return volatility risk from 10-K financial reports. The model provides explanations at word, sentence, and corpus levels, enhancing transparency and accountability in financial risk classification.
What it concludes
The study concludes that FinBERT-XRC enhances financial risk classification by providing transparent, multi-level explanations. Potential applications include improving investment decisions and regulatory compliance. Future research may focus on refining risk metrics and evaluating model explainability through expert assessments.
Evidence objects
The study concludes that FinBERT-XRC enhances financial risk classification by providing transparent, multi-level explanations. Potential applications include improving investment decisions and regulatory compliance. Future research may focus on refining risk metrics and evaluating model explainability through expert assessments.
key_findings bullet 1 · key_findings · validation V0
Raw abstract and provenance
Abstract: Every publicly traded company in the US is required to file an annual 10-K financial report, which contains a wealth of information about the company. In this paper, we propose an explainable deep-learning model, called FinBERT-XRC, that takes a 10-K report as input, and automatically assesses the post-event return volatility risk of its associated company. In contrast to previous systems, our pro… ▽ More Every publicly traded company in the US is required to file an annual 10-K financial report, which contains a wealth of information about the company. In this paper, we propose an explainable deep-learning model, called FinBERT-XRC, that takes a 10-K report as input, and automatically assesses the post-event return volatility risk of its associated company. In contrast to previous systems, our proposed model simultaneously offers explanations of its classification decision at three different levels: the word, sentence, and corpus levels. By doing so, our model provides a comprehensive interpretation of its prediction to end users. This is particularly important in financial domains, where the transparency and accountability of algorithmic predictions play a vital role in their application to decision-making processes. Aside from its novel interpretability, our model surpasses the state of the art in predictive accuracy in experiments on a large real-world dataset of 10-K reports spanning six years. △ Less
Source row: 754 · abstract type: unknown