March 13, 2017 16:30:15

Hierarchical Clustering

Scientists have observed the occurrence of two distinctive subsets in interplanetary coronal mass ejections (ICMEs): magnetic clouds (MCs) and non-magnetic clouds (non-MCs). We do not yet fully understand why some ICMEs are identified as MCs while others are not. There are features such as Plasma Speed, magnetic field magnitude etc. that suggest an ICME event to be an MC event, however it is far from an automated process. In addition to being time-consuming, the results differ depending on tightness of definition. In this paper we approach the MC and non-MC class distinction from a data analysis and data mining perspective and discover distinct patterns that set the two types of ICME events apart. We will be looking at the time series dataset from spacecraft Ulysses combined with a list of identified MC and non-MC data. The multivariate time series data will be hierarchically clustered with Dynamic Time Warping algorithm, and we will see how MC and non-MC events cluster and compare the results to the classifications made using traditional methods. This can potentially lead to automated identification of ICME events as well as predictions, which could prevent catastrophic events associated with solar activity from occurring.