The influence of weighting the k-occurrences on hubness-aware classification methods

Hubness is a phenomenon present in many highdimensional data sets. It is related to the skewness in the distribution of k-occurrences, i.e. occurrences of data points in k-neighbor sets of other data