Clustering
Cluster analysis or clustering is the process of finding meaningful groups (clusters) of points or objects in a dataset based on their similarity. As the result of this unsupervised data mining technique, we want points in each cluster to be similar to each other, while being different to points in other clusters.
Clustering of time-series is challenging because each data point is a period of time (an ordered sequence). It has found application in diverse areas to discover patterns that empower time-series analysis, extracting insights from complex datasets.
We are not going to get into details on time-series clustering, but the following table gives an overview of Python libraries for time-series clustering:
|
Library |
Implementations |
Maintainer |
Stars |
...
