Prerequisite: None

Course Outline

One of the striking things about machine learning is the wealth of techniques and algorithms available for modeling. There is seemingly a solution for every class of problem. It is easy to forget that data preparation is just as diverse. Each class of problem has a different optimal data structure associated with it. Techniques like Cluster Analysis, Association Rules, and Anomaly Detection – typically called Unsupervised Learning – are no different. They each have distinct data requirements.

It is not unusual to conflate unsupervised learning and supervised learning. A project can fail before it begins if the problem has not been properly defined. Carefully setting up the problem is a major theme of the course. For instance, cluster analysis is exploratory in nature and doesn’t generate a propensity score in the same way that supervised learning methods do. How do you take a model and automate it in support of organizational decision making without a propensity score? This course will show you how.

In this full day course, the instructor will demonstrate a variety of examples starting with the correct data structure for these methods. While not hands-on, well-known and widely available modeling algorithms will be demonstrated and discussed. Alignment between data format and problem definition will be reviewed. Different emphases of the problem will often imply modifications to the way that the data is prepared.

The course will then proceed with the exploration and interpretation of candidate models. Options for acting on results will be explored. You will also observe how a mixture of models including business rules, supervised models, and unsupervised models are used together in real world situations for various problems like insurance and fraud detection.

You Will Learn

• When to apply supervised or unsupervised modeling methods
• The importance of careful problem definition in the selection of a technique
• How to prepare data specifically for unsupervised methods
• Options for inserting unsupervised models into the decision-making process of the organization
• How to creatively combine supervised and unsupervised models for greater performance
• Effective techniques for deploying the results of unsupervised modeling
• Interpret and monitor your unsupervised models for continual improvement

Geared To

• Analytic Practitioners; Data Scientists; IT Professionals; Technology Planners; Consultants; Business Analysts; Analytic Project Leaders