Summary

The Segmentation and Clustering course provides students with the foundational knowledge to build and apply clustering models to develop more sophisticated segmentation in business contexts. You will learn:

• The key concepts of segmentation and clustering, such as standardization vs. localization, distance, and scaling

• The concepts of variable reduction and how to use principal components analysis (PCA) to prepare data for clustering models

• How to choose between hierarchical and k-centroid clustering models

• How to build and apply k-centroid clustering models

Throughout this course you’ll also learn the techniques to apply your knowledge in a data analytics program called Alteryx.

This course is part of the Business Analyst Nanodegree Program.

Expected Learning

Segmentation is used by companies across industries to better target the right products to the right customers. Most segmentation approaches are rather simple, such as segmenting customers by geography or age. However, with the rich amount of data business have now, much more sophisticated segmentation approaches are available.

In this course, you'll learn how to use an advanced analytical method called clustering to create useful segments for business contexts, whether its stores, customers, geographies, etc.

You'll learn this through improving your fluency in Alteryx, a data analytics tool that enables you prepare, blend, and analyze data quickly.

This course is ideal for anyone who is interested in pursuing a career in business analysis, but lacks programming experience.

Syllabus

Lesson 1 – Segmentation and Clustering Fundamentals

In this lesson, you’ll receive instruction and complete quizzes to help you understand concepts like standardization, localization, and distance. You’ll also see a few real world examples of how clustering and segmentation are used in business and other fields.

Lesson 2 – Data Preparation for Clustering Models

In this lesson you’ll learn how to select data based on the context of the business question. You’ll learn what data types can be used in clustering models, and how to scale and transform data in preparation for use in clustering models.

Lesson 3 – Variable Reduction

Clustering models can handle a lot of variables, but more variables can make interpreting results difficult. In this lesson, you’ll learn how and when to use a variable reduction technique called principal component analysis (PCA).

Lesson 4 – Clustering models design

In this lesson you’ll learn the important differences between the two main types of clustering models, k-centroid and hierarchical. You’ll learn how to decide how many clusters you should choose for your model, as well as how to validate your clusters in Alteryx.

Lesson 5 – Building a Clustering Model

In this lesson you’ll build a k-centroid clustering model to segment retail stores based on weather variables. You’ll learn how to visualize and validate your clusters. Lastly, you’ll learn how to interpret the results and communicate the "story" of the analysis.

Required Knowledge

• No programming experience required
• Interested in using data to make better business decisions
• Alteryx license (provided to Nanodegree students at no cost, compatible with Windows only)