Fabric notebook to do Customer Complaints Review

The repo has fabric notebook that loops through a Fabric Lakehouse table containing Customer Complaints from Consumer Financial Review Board. It sends the reviews to Azure OpenAI using the native OpenAI endpoint in Fabric and has it categorize the reviews based on a list of categories identified earlier while also providing sentiment.**

Using Azure OpenAI endpoints natively in Microsoft Fabric

A useful feature in Microsoft Fabric is the ability to use Azure AI services, like Azure OpenAI, directly in Fabric without having to deploy a separate endpoint or service. Per the Fabric documentation, "Fabric seamlessly integrates with Azure AI services, allowing you to enrich your data with prebuilt AI models without any prerequisite. We recommend using this option as you can utilize your Fabric authentication to access AI services, and all usage are billed against your Fabric capacity". This solution is currently in public preview.

Fabric AI Services Overview

Solution highlights:

• This repository will walk you through implementing a simple use-case leveraging the Fabric native Azure OpenAI endpoint. Note that you will need a workspace in a Power BI Premium capacity or a Fabric F-SKU (F64 and above) in order to use the built-in Azure AI services. The data in the solution is from an open dataset that contains Consumer Financial Review Board complaints data. See below link to download.

  Link to Consumer Financial Review Board complaints data

• The real-time pattern implemented here can be applied to expedite business insights & make faster business decisions.

  • Drive earlier actions: for example customer churn & upsell opportunities.
  • Live guidance to enable agent-assist, coaching and virtual agents use cases.

• Sentiment and insights from Complaints can be used to drive better customer experiences in effort to improve Customer Satisfaction (CSAT) scores.

  • Could be used as additional analytics by capturing real time stream from Customer Call Log too.

Key features of this solution:

1. Fabric Lakehouse: Load CSV file in Fabric Lakehouse and ingest directly into spark notebook.
2. GPT model within Fabric: Extract key category information & capture sentiment from customer complain narrative.
3. Extend Insights: Extract similar information from other data like company public response.
4. Reviews data Generates review topics and review sentiment that can be used in analytics.