期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII Part B2
页码：257-262
出版社：Copernicus Publications
摘要：The usage of single geographical footprint model in existing Geographical Information Retrieval (GIR) system will cause many problems like overestimation or underestimation of the geographical scopes for documents to search. To be honest, the single geographical footprint model is not applicable in modern GIR system although it is simple, fast and widely applied today. In order to improve the quality of answers given to a spatial query, a new model as well as a dedicated algorithm will be proposed to study the geographical information attached to documents. Besides, a dedicated algorithm based on network graph, which is inspired by the Google PageRank and Bayesian network theory, is also invented to estimate the geographical similarity between document € d and query € q based on this new model. We believe that the new model and algorithm proposed here could better estimate the geographical similarity between document and query for a GIR system. Because it not only consider the geometric adjacency of document and query, but also take into account the importance of each places towards the hosting documents. By using the new model, the geographical scope of document could be better studied, and thus improve the quality of answers from GIR system concerning the spatial query
关键词：Spatial Information Science; GIS; Information; Service; Information Retrieval; Algorithm; Retrieval M odel
