Using minimaps to enable toponym resolution with an effective 100% rate of recall

Abstract

A number of systems have been recently constructed that make use of a map query interface to access documents by the locations that they mention. These mentions are often ambiguous in the sense that many interpretations exist for the locations which are not always expressed along with all the necessary qualifiers. In other words, users are assumed to be able to make the appropriate identification based either on knowledge of prior queries or the nature of the document containing the references as well as knowledge of the target audience. The disambiguation process is known as toponym resolution. The map query interface results in the placement of icons and links to the appropriate documents at the corresponding location on the map. Assuming that all toponyms have been recognized (i.e., 100% rate of recall for toponym recognition), it is shown how to achieve an effective 100% rate of recall for toponym resolution for all interpretations of a toponym that the toponym recognition process associates with at least one document. This is done with the aid of a minimap that shows all of these interpretations which means that a user has access to all documents that mention a specific location as long as the textual specification to the location has been recognized as a location rather than as the name of another entity such as a person, company, organization, etc. It also assumes that the user is capable of determining the correct interpretation of each toponym. This is important as it enables the determination of precision and recall.

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 — 19 citations exist upstream.

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