STEWARD: demo of spatio-textual extraction on the web aiding the retrieval of documents

ABSTRACT

A spatio-textual search engine, termed "STEWARD" is demonstrated where document similarity is based on both the textual similarity as well as the spatial proximity of the locations in the document to the spatial search input. STEWARD's performance is enhanced by the presence of a document tagger that is able to identify textual references to geographical entities. The user-interface of STEWARD provides the ability to browse results, thereby making it a valuable "knowledge discovery" tool.

REFERENCES

• 1: Hanan Samet , Houman Alborzi , František Brabec , Claudio Esperança , Gísli R. Hjaltason , Frank Morgan , Egemen Tanin, Use of the SAND spatial browser for digital government applications, Communications of the ACM, v.46 n.1, January 2003 [doi>10.1145/602421.602453]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

General Terms:
Algorithms, Design, Performance

Keywords:
STEWARD, geocoding, spatio-textual search engine