Energy Credit Best Practices – Chapter: Information Technology http://ccro.org © Copyright 2022, CCRO. All rights reserved. 30 3.5.4 Natural Language Processing (NLP) A sub-field of AI, NLP facilitates a “natural” way of communicating between humans and computers/machines. While there are many options for the deployment of an NLP system, the most common implementation is in the form of “Chatbots. These virtual assistants mimic a human personality and attempt to answer common questions that end users may ask. From a business perspective, this alleviates people's need to engage either online or via phone directly. Additionally, it minimizes exception processing, improves the quality of data captured, and consistently applies business rules to any conversation. Much like other advanced IT processing of data, the quality of interaction is directly driven by the amount and quality of data that these virtual assistants have access to “learn.” 3.5.5 Digitization of Credit Process The Lack of The Credit Process's digitization is tremendously risky in itself. While many actions and activities have moved into the digital world, for technologies such as AI, ML, and others to provide the most valuable, all aspects of the Credit Process must be digitized. Often, exceptions and other one-off processes are handled manually, and thus their digitization – and the logic that formulated those decisions – are not “learned” by the Advanced IT techniques. 3.5.6 Conclusion Advanced IT techniques can help credit professionals draw insights from vast volumes of market data and support credit decisions more than traditional analog approaches. Energy companies can no longer exclusively rely on humans to establish direct parameters & rules for algorithms or other data-intensive activities. Allowing Advanced IT techniques to derive more profound insights into your data while the credit analyst remains responsible for credit valuation and mitigation.