Geospatial Search: Building rich features for validation, analysis and search of geospatial data
Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
The main scope of this work is to develop enhanced search features over geospatial data in order to improve the user experience. The problem consists in finding a way to enable the entire geospatial dataset to be searchable. The goal is to add a new search parameter to the pre-existing full-text search query so that geospatial data is taken into consideration and better satisfaction of users needs is achieved. The report gives a clear theoretical overview of spatial data structures, while in the meantime relating their characteristics to various commercial applications, in the hope for the reader to use it as a reference and operate better informed choices with regards to geographical applications.
The method followed is principally the comparison between different solutions through the entire course of the work. Any single choice is weighted after thorough comparison of existing options, reading of papers, benchmarking or expert's opinions.
Heuristics to assess quality of data and to achieve validation of data are created. The principal chosen solution is selected among a few commercial options and eventually enables the geographical data to be searchable. A Python web-service implementation is described in order for the search features to be accessible to the end users.
Experiments run on production code are presented to demonstrate the efficacy of the implemented heuristics. Benchmarking experiments show the validity of the solution for geo-searching.
Place, publisher, year, edition, pages
IdentifiersURN: urn:nbn:se:kth:diva-165082OAI: oai:DiVA.org:kth-165082DiVA: diva2:807187
Subject / course
Master of Science - Computer Science
2015-04-13, 20:52 (English)