Intro
Something you may have noticed about all of the previous sections is the need for data to run the models we described. Most problems involve using data in order to predict something new, however there are also models that do not need data to work. This is called unsupervised learning. As opposed to the supervised learning of past models, it doesn’t require human “supervision” by providing data. In this section you will learn to apply unsupervised models to solve problems without prior data.
Application
The problem we will be going through for this section has to do with detecting anomalies. For example, say you were given a dataset but in your data exploration phase you see outliers or data that just doesn’t fit. Through an unsupervised model we can sort these outliers out to clean our dataset and ignore the human error possible when making datasets. Before we can apply this we need to first understand how the model will see the data.
Add a page link block here (if there is no previous section, delete this column)
Add a page link block here (if there is no next section, delete this column)
Copyright © 2021 Code 4 Tomorrow. All rights reserved.
The code in this course is licensed under the MIT License.
If you would like to use content from any of our courses, you must obtain our explicit written permission and provide credit. Please contact classes@code4tomorrow.org for inquiries.