Cluster-based dual evolution for multivariate systems
share
This paper proposes a cluster-based method to analyse multivariate systems that change over time. At each point in time, we partition data points into an appropriate number of clusters and thereby track both the total number of clusters and the individual constituents' cluster memberships changing over time. We can also compare and contrast several multivariate systems and determine patterns and anomalies between different data. We apply this method to study the evolution of COVID-19 cases and deaths over time. We detect significant similarities in the evolution of cases and deaths up to a suitable offset in time. We compute this offset and use it to identify anomalous countries in the progression of cases to deaths.
READ FULL TEXT