Unsupervised learning introduction
These algorithms are used where there is no target or outcome variable to predict, and the primary aim is to identify clusters of items in a dataset according to specific features or characteristics. These algorithms typically partition the data based on features similarities and differences to produce clusters. The computer learns to identify these data patterns without human guidance about how the different clusters should be determined. The input data is typically termed as being “unlabelled”, meaning they do not have any prior annotation that can be used to guide the partitioning.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.