Unsupervised segmentation of a univariate time series dataset using motifs and shapelets

A computer-implemented method comprising:

• receiving a first time series dataset from a sensor of a first monitored processor;
• grouping the first time series dataset into a plurality of clusters;
• using an unsupervised machine learning model, combining a subset of the plurality of clusters by data signature similarities to form a plurality of motifs;
• combining the plurality of motifs into one or more shapelets in accordance with the data signature similarities
• predicting, using a supervised machine learning model, a second time series dataset using the plurality of motifs or the shapelets;
• and
• initiating an action based on the prediction of the second time series dataset, wherein the action comprises:
• identifying an overutilization of at least one of the first monitored processor and a second monitored processor different than the first monitored processor;
• and
• transmitting an electronic communication that indicates the overutilization of the at least one of the first monitored processor and the second monitored processor.

A non-transitory computer-readable storage medium storing a plurality of instructions executable by one or more processors, the plurality of instructions when executed by the one or more processors cause the one or more processors to:

• receive a first time series dataset from a sensor of a first monitored processor;
• group the first time series dataset into a plurality of clusters;
• using an unsupervised machine learning model, combine a subset of the plurality of clusters by data signature similarities to form a plurality of motifs;
• combine the plurality of motifs into one or more shapelets in accordance with the data signature similarities;
• predict, using a supervised machine learning model, a second time series dataset using the plurality of motifs or the shapelets;
• and
• initiate an action based on the prediction of the second time series dataset, wherein the action comprises:
• identifying an overutilization of at least one of the first monitored processor and a second monitored processor different than the first monitored processor;
• and
• transmitting an electronic communication that indicates the overutilization of the at least one of the first monitored processor and the second monitored processor.