06:29 - 09:29
3m
Illustrates how once-separate diagnoses were placed together under the diagnosis of autism spectrum disorder thanks to heirarchical clustering, resulting in better treatment regimens.
Video Transcript
Related Clips
Explains how unsupervised machine learning's purpose is to create new groups to put data into, and the clustering types used in this process.
An example of how a pizza restaurant would use the k-means method of machine learning and centroids to find new groups of customers.
Explains how the silhouette method evaluates unsupervised machine learning processes by measuring the closeness of data points.
Explains how heirarchical clustering finds smaller subgroups within larger differentiated clusters, and how these groups can be visualized on a dendogram.
Gives an example of using a neural network to predict one's salary based on a number of different characteristics and by using an activation function.