Quarterly Mark December 2023 - 16

EXHIBIT 7
Example of Some Risk Factors That Are Learned from 10-K Documents Using a Topic Modeling Algorithm
Along with the Words That Contribute to the Risk Topics
Risk Category
Debt Risk
Drug Research Risk
Government Funding Risk
Advertising Risk
Risk Topic Words
Debt, Payment, Acquisition, Default
Clinical Trial, Marketing, Approve
State, Healthcare, Medicare
Station, Advertising, Broadcast
Example Companies
HEE Equipment Services;
US Concrete, Inc.
Hepion Pharmaceuticals,
Leap Therapeutics
Chemed Corp.; US Physical
Therapy, Inc.
Beasly Broadcast Group,
Sinclair Broadcast
NOTES: The third column represents the companies that are exposed to the respective risk factors. Using text-based risk factors, we can
find the exposure of different risk factors for each company. For example, Lopez-Lira (2023) found that Apple, Inc. is exposed to supply
chain risk, software risk, international risk, internet risk, and demand risk. Such fine-grained risk automated analysis from the textual data
is possible due to NLP models.
SOURCE: The source of risk factors is Lopez-Lira (2023).
for financial sentiment tasks. Today, NLP methods from rules-based to deep NLP
models are widely used in finance for a variety of tasks, from sentiment analysis to
risk management and algorithmic trading, among others.
A detailed literature review of existing works pertaining to financial NLP
is beyond the scope of this article, but in-depth surveys can be found in Xing,
Cambria, and Welsch (2018); Fisher, Garnsey, and Hughes (2016); and Kantos et al.
(2022). We discuss some of the most important NLP applications in the next few
subsections.
Risk Management
Risk management in finance refers to the process of identifying, assessing, and
controlling risks in financial operations, transactions, and investments. It involves
identifying potential sources of financial loss and taking steps to minimize or
mitigate their impact on the financial stability and performance of an organization.
Key components of financial risk management include market risk, credit risk,
operational risk, liquidity risk, and reputational risk. The goal of financial risk
management is to minimize the impact of adverse events on an organization's
financial performance and stability. It is one of the foremost applications of NLP in
finance.
NLP methods are used to extract data from news articles, financial reports, and
regulatory filings to identify patterns and relationships among risk factors. Healy and
Lo (2011) performed language analysis to identify event risk, and Buehlmaier and
Whited (2018) developed a method to determine risk factors via textual analysis.
Exhibit 7 shows some example of risk factors obtained using topic modeling. NLP
can also be used for regulatory compliance to mitigate financial risk, as it can be
used to automatically scan financial documents and identify potential compliance
violations, such as insider trading or money laundering. Zhang and El-Gohary
(2016) discussed an NLP-based automated compliance checking system. For credit
16 | From ELIZA to ChatGPT: The Evolution of Natural Language Processing and Financial Applications

Quarterly Mark December 2023

Table of Contents for the Digital Edition of Quarterly Mark December 2023