Smart Data Discovery Using SAS Viya

Chapter 7: Machine Learning and Visual Modeling

Introduction

Why Visual Modeling

Approaches and Techniques

Overall approach

Techniques and Algorithms

Preparing the Data for Modeling

Data Partitioning

Dealing with Missing Data

Feature Engineering

Outlier Processing

Binning

Log Transform

Splitting/extracting features

Normalize/standardize variables

Model Assessment

Variable Importance

Model Fit Statistics

Misclassification Chart

ROC Chart

Lift Chart

Improving Your Model

Avoid Under-fitting and Over-fitting

Adding a Group-By Variable

Using Better Hyperparameters

Using Your Model

Introduction

“The signal is the truth. The noise is what distracts us from the truth.”

– Nate Silver

Machine learning can be loosely defined as using a variety of supervised ...