Identifying Demographic Changes in Europe: Empirical Application of Soft Clustering

Absztrakt (kivonat)

Our goal is to identify demographic characteristics of European regions (NUTS–2) focussing on fertility and mortality rates, life expectancy, and female employment rate. Traditionally, sorting areas into groups is done by clustering. Clustering is an unsupervised machine learning technique that divides the population into several groups in which points in the same group are similar to each other and points in different groups are dissimilar. Repeating the algorithm in each year would cause confusing jumps in the cluster centres. Thus, instead of putting each data point into separate clusters, the probability of that point being in that cluster is assigned. In soft clustering or fuzzy clustering, each data point can belong to multiple clusters along with its probability score or likelihood.