Course Description
Machine Learning is an application within artificial intelligence that consists of a collection of models, methods and algorithms that find patterns in big data and help organizations make better business decisions. Decisions made using machine learning are driven by data instead of hunches or suspicions. This is the second certificate in a series of Applied certificates on Machine Learning for Artificial Intelligence. In this series, students will gain a practical understanding of the tools and techniques used in Machine Learning. This is the second part in a series of courses on Applied Machine Learning and Artificial Intelligence. In this course, students are introduced to supervised machine learning regression methods and algorithms used in Artificial Intelligence. Students will learn to use Cloud Technologies, Python and Jupyter Notebook to train and deploy supervised machine learning models built using regression algorithms such as Linear Models (OLS), Ridge and Lasso, Neural Networks, Ensemble and Support Vector Machines for forecasting sales volume, estimating solar irradiance, predicting credit limits and estimating energy usage. Students must complete all 5 modules to complete the certificate.Course Outline
Introduction
Module 1: Using Ordinary Least Squares Regression to study the relationship between nutrition, physical activity and obesity
In this task, students will use a parametric dataset and build Linear Models to study relationships and trends in obesity from the Behavioral Risk Factor Surveillance System in the United States.
Objectives:
1. Introduction to Regression
2. Correlation
3. Collinearity
4. Assessing Regression Models
Module 2: Predicting sales volume with Gradient Tree Boosting ensemble learning
In this task, students use sales data and feature descriptions from similar products on the market and predict the first-year sales volume for a new marketing scenario.
Objectives:
1. Principal Component Analysis
2. Covariance
3. Loss
4. Learning rate
5. Splitting criterion
Module 3: Predicting Solar Irradiance with Neural Networks
In this task, students will build a local solar prediction model that will be used to control the efficiency of local solar microgrids.
Objectives:
1. Multi-Layer Perceptrons
2. Backpropagation
3. Layer Design
4. Activation Functions
5. Deep Learning
Module 4: Automating loan approvals based on customer payment history and demographics using Support Vector Regression
In this task, students work for a credit underwriting service that needs to build an automated system for improving the accuracy and efficiency of approving consumer loans.
Objectives:
1. Kernels and degrees
2. Learning rate (C)
3. Epsilon
4. Deployment with Flask
Module 5: Predicting energy consumption with CART algorithms
Using a large composite of different recorded energy readings from submeters, students will build a predictive model that can be used to predict energy consumption based on different weather-related features and consumer actions.
Objectives:
1. Understanding CART models
2. Using SQL to retrieve data
3. Sampling data with simple sampling
4. Training metrics
Learner Outcomes
- Using machine learning tools to investigate patterns in complex data sets
- Preprocessing data for machine learning tasks
- Using statistical machine learning to investigate regression problems and correlated data
- Using Ensemble Learning to investigate regression problems
- Assessing the predictive performance of predictive models
- Drawing relationships between learner performance and measured features to help understand model performance
- Conducting feature selection to investigate the correlation between different features in a dataset
- Presenting machine learning results to a non-technical audience