Machine LearningLesson 87Free preview

K-means

This lesson is part of Python · All levels

Summary

K-means is an unsupervised learning algorithm that groups data into k clusters based on their proximity to the centroid. How do you do it: determine the number of k clusters, place the centroid randomly, then re-assign the data to the nearest centroid and recalculate the centroid until it's stable. The number of k clusters is a hyperparameter that must be determined before training, often aided by the elbow method. K-means are widely used for customer segmentation and data grouping without labels.

Key points

K-means is an unlabelled clustering algorithm.
The number of k clusters must be determined before the model is trained
Iteration: data to the nearest centroid and then the centroid is recalculated
The elbow method helps to select the right number of clusters

Full lesson

Code examples and quizzes

Full lessons, code examples, and quizzes are available to subscribers. 5 questions.