or

This curriculum module teaches students basic machine learning algorithms and how to apply them to biological datasets.

Students will explore different types of machine learning and related algorithms by looking at two datasets: patient ovarian cancer data and data about mollusk shell sizes. Students will need to utilize the Statistics and Machine Learning Toolbox™ for this module. This module assumes basic MATLAB knowledge and we recommend that all students take the MATLAB Onramp before continuing if they have not already. Students should also review the biosciences module on statistics, as some of the concepts introduced there are used in this module.

A computer building a model to differentiate pictures of dogs from pictures of cats is a classic machine learning example.

This module assumes basic MATLAB knowledge and it is recommended that all students take the MATLAB Onramp. Students should have previously completed the biosciences data module and the biosciences statistics module.

To learn more about opening and using MATLAB, see the accompanying Getting Started guide.

Notes: These scripts can all be run independently, though we recommend going through these live scripts in order. These live scripts are intended to be used with output inline. To change the output, go to the View tab of the toolstrip, and select Output Inline. The scripts have areas for the students to interact with the code . There will also be exercises in most scripts and the answers will be provided at the end. A problem set for students to practice these concepts is also included here. Throughout the scripts, there are also moments to students to reflect on what they've learned or on what the data means . Particularly interesting examples of how these concepts are used in "real-world" biology are also pointed out .

Introduction

• Learning objective: Students will get a brief introduction to machine learning and how it is used in biology.

Unsupervised learning

• Learning objective: Students will learn techniques such as dimensionality reduction and clustering and apply them to analyze patient ovarian cancer data.

Supervised learning

• Learning objective: Students will learn to use MATLAB apps to apply classification and regression learning techniques to both patient and ecological data.

Problem set

• Problem set answers

Link to 5 other modules here once set up.

Students may also want to work through the Machine Learning Onramp.

MATLAB®, Statistics and Machine Learning Toolbox™

The License for this project is in the License.txt file in this repository.

© Copyright 2023 The MathWorks, Inc.