Clustering of COVID-19 Multi-Time Series-Based K-Means and PCA With Forecasting

2. Related Work

To more effectively locate the first cluster centroids, a number of methods have been suggested. In the Encyclopedia of Ecology, Craig Syms, Principal Components Analysis, its main function is to display the relative locations of data points in fewer dimensions while preserving the most significant amounts of information that can explore relationships between dependent variables. Olive (2017) used classical PCA in explaining the reduction and concentration structure with a few linear uncorrelated combinations of the original variables. He implemented PCA in data reduction and interpretation. In relation to the Walk-Random hypothesis, previously rejected in many previous studies, the current study showed a decrease in the time-series correlation at least in part of the Japanese stock market - which means that the models that Dependent on that hypothesis have become more important (Cao et al., 2015).