Enhancing spatial query results using semantics and multiplex networks
Links to Fileshttp://heteronam.org/2018/papers/HeteroNAM_2018_paper_2.pdf
MetadataShow full item record
Type of Work8 pages
conference papers and proceedings
RightsThis item may be protected under Title 17 of the U.S. Copyright Law. It is made available by UMBC for non-commercial research and education. For permission to publish or reproduce, please contact the author.
SubjectsGeographic information systems
dynamic graph algorithm
Geographic information retrieval
Geographic information retrieval (GIR) research mainly focuses on ﬁnding relevant geographic artifacts along one layer, although answers to queries should preferably include artifacts across multiple layers.This paper addresses this limitation, by exploiting relationships that exist between geographical objects of the same or diﬀerent types. Given the existence of these relationships we can infer additional indirect semantic relationships between geographical objects that otherwise would be ignored. This semantic information contains multiple dimensions. This research creates a mathematical model of semantically relevant geographic objects that belong to diﬀerent layers using the construct of multiplex networks. The proposed model respects and extends the “layered” structure of geographic data, and enhances the results of queries made to a GIS. A prototype system has been implemented over an existing GIS relational database. Experiments with the prototype show signiﬁcant improvements in the results of queries, measured using precision and recall.