This course combines data exploration, visualization, data preparation, feature engineering, sampling and partitioning, model training, scoring, and assessment.
It covers a variety of statistical, data mining, and machine learning techniques performed in a scalable and in-memory execution environment.
The course provides theoretical foundation and hands-on experience with SAS Visual Data Mining and Machine Learning through SAS Studio, a user interface for SAS programming.
The course includes predictive modeling techniques such as linear and logistic regression, decision tree and ensemble of trees (forest and gradient boosting), neural networks, support vector machine, and factorization machine.
Who should attend:
Data analysts, data miners, mathematicians, statisticians, data scientists, citizen data scientists, qualitative experts, and others who want an introduction to supervised machine learning for predictive modeling
Before attending this course, you should have, at minimum, an introductory-level familiarity with basic statistics. SAS experience is helpful but not required. Coding experience is helpful but not required.
This course addresses SAS Viya software.
Delegates will learn how to
Learn how to:
- Create a SAS Cloud Analytic Services (CAS) session, and prepare and explore data for machine learning.
- Build linear and logistic regression models.
- Build decision tree, forest, and gradient boosting models.
- Build neural network models.
- Build support vector machine models.
- Build factorization machine models.
- Evaluate and compare model results.
- Score selected models
Introduction to SAS Viya, Data Preparation, and Exploration
- Introduction to SAS Viya.
- SAS Cloud Analytic Services architecture.
- SAS Cloud Analytic Services sessions.
- SAS Cloud Analytic Services libraries.
- Development environment: SAS Studio.
- Data preparation and exploration.
- Concepts of predictive modeling.
- Introduction to regression.
- Managing missing values.
- Selecting regression inputs.
- Optimizing regression complexity.
- Interpreting regression models.
- Transforming inputs.
- Categorical inputs.
- Polynomial regressions.
- Tree-structure models.
- Recursive partitioning.
- Ensemble of trees.
- Neural Network
- Network architecture.
Support Vector Machine
- Large-margin linear classifier.
- Methods of solution.
- Nonlinear classifier: kernel trick.
Model Assessment and Scoring
- Model assessment and comparison.
- Model deployment.
Factorization Machines (Self-Study)
- Factorization machines.