Databricks Insurance Q&A NLP

Digitalization of the Claims Process using NLP, Delta and Hugging Face

While companies across industries have accelerated digital adoption, are insurers meeting the ever-changing demands from customers? As an insurer, are you spending most of your time and resources into creating business value?

Customer service is a vital part of the insurance business, for multiple business cases, from marketing, to customer retention and claims. By digitalizing these processes, insurers can seamlessly scale their operations and shift money and resources from operational processes to actual product and value creation.

This solutions accelerator is a head start on developing and deploying a machine learning solution to detect customer intents based on pieces of text from an Interactive Voice Response (IVR) stream or from a virtual agent.

Target Solution

Requirements

Please use the following package and environment versions when running this on Databricks:

• Databricks Machine Learning Runtime 12.0
• pytorch_lightning==1.8.6
• transformers==4.23.1

Authors

• Rafael Piere (rafael.pierre@databricks.com)