On the Provenance of Linked Data Statistics

TL;DR

This paper discusses the challenges in consistently describing and interpreting statistics of linked data graphs, emphasizing the need for clear provenance to understand how such statistics are calculated.

Contribution

It surveys the problem of expressing and understanding linked data statistics and proposes a strategy to address the issue of provenance and interpretability.

Findings

Highlights the difficulty of uniform statistical descriptions

Proposes a strategy for provenance tracking in linked data statistics

Emphasizes the importance of understanding calculation methods

Abstract

As the amount of linked data published on the web grows, attempts are being made to describe and measure it. However even basic statistics about a graph, such as its size, are difficult to express in a uniform and predictable way. In order to be able to sensibly interpret a statistic it is necessary to know how it was calculate. In this paper we survey the nature of the problem and outline a strategy for addressing it.