[statistics] One divided by distance, often raised to some power (1/D or 1/D2, for example), where D is a distance value. By inverting the distance among spatial features, and using that inverted value as a weight, near things have a larger weight or influence than things that are farther away.