Introduction to machine learning: Clustering
00:00 Intro
00:13 Contents
00:40 Recap
3:33 What is clustering?
4:03 Why use it?
5:57 Clustering applications
8:10 What is a cluster?
9:29 Types of clustering algorithms
13:30 KMeans algorithm
19:47 KMeans initialisation
22:27 KMeans evaluation methods
25:48 KMeans strengths and weaknesses
28:49 Hierarchical clustering
32:56 Types of hierarchical clustering
33:58 Measure of distance
36:15 Linkage criterion
37:53 Example
43:35 Strengths and weaknesses
What is machine learning? How is machine learning different from classic statistics? What are its applications? What type of models exist within machine learning?
If these are questions that you have then come along to this free three-part webinar that has been designed to deepen your understanding of the main concepts present within machine learning. Machine learning combines statistics and computer science to draw inference from patterns in data. It has now become an indispensable skill for data scientists and statisticians alike. These webinars will explore a few of the most important machine learning algorithms and then discuss model selection and evaluation of these models.
This second session moves on to exploring a specific unsupervised method, clustering. We will cover the following types of clustering algorithms:
centroid-based: specifically, k-means algorithm
hierarchical-based: divisive (top-down) and agglomerative (bottom-up)
You can find the resources on the event web page: https://ukdataservice.ac.uk/events/in...