Introduction to More Use Cases


These use cases cover the following concepts and features in Dataiku DSS:


  1. Narrow and wide data structures
  2. Spatial joins
  3. Feature generation
  4. Supervised learning
  5. Unsupervised learning
  6. Model retraining
  7. Schema and Types
  8. Projects
  9. Plugins

Concepts unique to Dataiku or uniquely implemented on the platform are italicized.


  1. File import, Schema definition
  2. Visual Recipes: Download, Prepare, Pivot, Group, Join, Window
  3. Visual Machine Learning tool: classification, clustering
  4. Explore & Analyze tool
  5. Charts, dashboards, web apps
  6. Plugins: Reverse Geocoding, Geocoder, Get US census block

The learning objectives of these use cases together are:

  • conduct data preparation and feature generation, including spatial joins and geocoding through plugins
  • carry out Exploratory Data Analysis using descriptive statistics, charts, visualizations and samples
  • create supervised and unsupervised models using visual wizards and retrain them using new data


These use cases assume knowledge of the platform covered in the tutorials. Please check that you are familiar with the following tutorials before starting the use cases:

Technical Requirements

The use cases are implemented in Dataiku DSS, so the basic requirement is access to an instance of the platform where you can create projects (or have an administrator create projects for you).

In addition, some of them require plugins. Here is the complete list of plugins used; ask an administrator to install them.

Wrap up

Let’s get started!