Use case tutorials
Take use case tutorials to experience one or more of the use cases that combine to demonstrate how you can implement a solution with Cloud Pak for Data. The tutorials follow the story of Golden Bank, a leading mortgage provider that needs to solve the challenges of data access, data quality, data governance, and managing data and AI lifecycles.
If you want to learn how to complete a specific task, such as refine data or build a model, then consider taking the quick start tutorials. The quick start tutorials help you quickly learn how to do a specific task or set of related tasks.
The use cases tutorials are categorized by type:
- Data fabric tutorials show you how to implement one of the data fabric use cases.
- Build and govern AI tutorials show you how to build, deploy, and monitor models in a governed environment.
Data fabric tutorials
A data fabric is an architecture that provides a secure and consistent way to access data from disparate sources and a set of integrated tools so that your organization can efficiently collaborate to use your data to improve your business.
Watch this video to see data fabric in action.
This video provides a visual method to learn the concepts and tasks in this documentation.
The data fabric is split into use cases that each represent a particular goal. On IBM Cloud Pak for Data, services provide features and tools. Each use case requires one or more service instances. Some services are included in multiple use cases.
The tutorials are grouped by use case. You can start with any use case. Each group of tutorials is based on a sample project that contains the resources that you need to complete the tutorials. You download a sample project from the link in the tutorial and then import that sample project.
The tags for each tutorial describe the level of expertise (, , or ), the amount of coding required ( or ), and whether the tutorial is a continuation () of one or more other tutorials that you must complete first.
Data integration
Provide access to all your data, without moving it.
Scenario: Golden Bank needs a data pipeline that delivers concise, pre-processed, and up-to-date data on all mortgage applicants, so that lenders can make decisions.
Click a tutorial for this use case to get started:
- Integrate data
Extract, filter, join, and transform your data.
- Virtualize external data
Virtualize and join data tables from external sources.
- Orchestrate an AI pipeline with data integration
Create an end-to-end pipeline that transforms data and trains a model.
Data governance
Share, enrich, and govern data.
Scenario: Golden Bank needs to create a business vocabulary to describe and manage data assets, and then make those assets available in a self-service catalog.
Click a tutorial for this use case to get started:
- Curate high quality data
Create high quality data assets by enriching your data and running data quality analysis.
- Protect your data
Control access to data in a catalog.
- Consume your data
Evaluate, share, shape, and analyze data.
- Govern virtualized data
Enrich virtualized data and ensure that virtual data is protected.
Master Data Management
Provide a single, trusted 360-degree view of your data.
Scenario: Golden Bank wants to run a campaign to offer lower mortgage rates. As a data engineer, you must use IBM Match 360 to set up, map, and model your data for a 360-degree view of the customer.
Click a tutorial for this use case to get started:
- Configure a 360-degree view
Set up, map, and model your data to create a 360-degree view of your customers.
Build and govern AI tutorials
These Cloud Pak for Data use cases cover the tools to build and govern AI. You can start with any use case. Each group of tutorials is based on a sample project that contains the resources that you need to complete the tutorials. You download a sample project from the link in the tutorial and then import that sample project.
The tags for each tutorial describe the level of expertise (, , or ), the amount of coding required ( or ), and whether the tutorial is a continuation () of one or more other tutorials that you must complete first.
AI governance
Build, operationalize, and govern AI.
Scenario: Golden Bank needs a model that identifies whether customers qualify for mortgages to reduce the bank's application processing costs.
Click a tutorial for this use case to get started:
- Build and deploy a model
Train a model, promote it to a deployment space, and deploy the model.
- Test and validate the model
Evaluate a model for accuracy, fairness, and explainability.
Data Science and MLOps
Build, deploy, and monitor models.
Scenario: Golden Bank needs to automate a data pipeline that delivers up-to-date data on all mortgage applicants, that lends can use for decision making.
Click a tutorial for this use case to get started:
- Orchestrate an AI pipeline with data integration
Create an end-to-end pipeline that transforms data and trains a model.
- Orchestrate an AI pipeline with model monitoring
Create an end-to-end pipeline that prepares data and trains a model, and then use Watson OpenScale to validate the model.