Probabilistic Approach for Modeling and Presenting Error in Spatial Data

Abstract

This paper presents a probabilistic approach to describe and visualize uncertainty and error in spatial data. The probabilistic approach assigns n-dimensional probability zones to n-dimensional measured feature locations. The size of each n-dimensional zone depends on the uncertainty arising from imprecise measurements or derived inaccuracy values, and a user-selected probability threshold that the “true” feature location is to be found within this probabilistic space. The uncertainties relate to recorded measurement precision, accuracy of a linear network, and issues of scale and resolution. The confidence intervals are based on the χ2 distribution, where the probability that the measured point location is within the tabulated distance of the true point location can be tested. The error model computes the probability of the intersection of two features or data sources to determine whether they are compatible and if they should be used together. The error model allows the user to assess the potential quality implications of combining data from different sources and with different qualities. The error model is encapsulated in a software program that includes a graphic user interface that facilitates visualization of results. The ability to visualize the quality of spatial data at different significance levels of confidence provides a powerful tool for communicating the impacts of the quality of spatial data on applications of interest to users.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.