Machine Learning Classification Algorithms using MATLAB [Video]
-
Free ChapterInstructor and Course Introduction
-
MATLAB Crash Course
-
Grabbing and Importing a Dataset
-
K-Nearest Neighbor
- Nearest Neighbor Intuition
- Nearest Neighbor in MATLAB
- Learning KNN model with features subset and with non-numeric data
- Dealing with scaling issue and copying a learned model (4)
- Types of Properties (5)
- Building a model with subset of classes, missing values and instances weights (6)
- Properties of KNN
-
Naive Bayes
-
Decision Trees
-
Discriminant Analysis
-
Support Vector Machines
-
Error Correcting Output Codes
-
Classification with Ensembles
-
Validation Methods
-
Performance Evaluation
- Making Predictions with the Models
- Determining the classification loss
- Classification Margins and Edge
- Classification Loss, Margins, Predictions and Edge for cross validated models
- Comparing two classifiers with holdout
- Computing Confusion Matrix
- Generating ROC Curve
- Generating ROC Curve based on the testing data
- More Customization and information while generating ROC
- Computing Accuracy, Error Rate, Specificity and Sensitivity (10)
This course is designed to cover one of the most interesting areas of machine learning called classification. I will take you step-by-step in this course and will first cover the basics of MATLAB. Following that we will look into the details of how to use different machine learning algorithms using MATLAB. Specifically, we will be looking at the MATLAB toolbox called statistic and machine learning toolbox. We will implement some of the most commonly used classification algorithms such as K-Nearest Neighbor, Naive Bayes, Discriminant Analysis, Decision Tress, Support Vector Machines, Error Correcting Output Codes and Ensembles. Following that we will be looking at how to cross validate these models and how to evaluate their performances. Intuition into the classification algorithms is also included so that a person with no mathematical background can still comprehend the essential ideas. The following are the course outlines.
Segment 1: Instructor and Course Introduction
Segment 2: MATLAB Crash Course
Segment 3: Grabbing and Importing Dataset
Segment 4: K-Nearest Neighbor
Segment 5: Naive Bayes
Segment 6: Decision Trees
Segment 7: Discriminant Analysis
Segment 8: Support Vector Machines
Segment 9: Error Correcting Output Codes
Segment 10: Classification with Ensembles
Segment 11: Validation Methods
Segment 12: Evaluating Performance
Style and Approach
This course is really good for a beginner. It will help you to start from the ground up and move on to more complicated areas. You receive knowledge from a Ph.D. in Computer science (machine learning) with over 10 years of teaching and research experience,
- Publication date:
- December 2017
- Publisher
- Packt
- Duration
- 6 hours 53 minutes
- ISBN
- 9781788992213