Advanced

A recommender geoportal for geospatial resource discovery and recommendation

Dareshiri, Shokouh; Farnaghi, Mahdi LU and Sahelgozin, Mohammadreza (2019) In Journal of Spatial Science 64(1). p.49-71
Abstract

Geoportals have been widely implemented by national spatial data infrastructures (NSDI) to provide users with the ability to search for geospatial resources over the Internet. This paper proposes an approach that can improve the functionalities of the conventional geoportals by adding recommendation capabilities to the execution mechanisms of geoportals. The geoportals with recommendation capability are able to analyse users’ activities and recommend geospatial resources to users based on their desires and preferences. A hybrid recommendation mechanism based on the singular value decomposition (SVD) method is developed to provide required prediction and recommendation capabilities. Spatial attributes of geospatial resources along with... (More)

Geoportals have been widely implemented by national spatial data infrastructures (NSDI) to provide users with the ability to search for geospatial resources over the Internet. This paper proposes an approach that can improve the functionalities of the conventional geoportals by adding recommendation capabilities to the execution mechanisms of geoportals. The geoportals with recommendation capability are able to analyse users’ activities and recommend geospatial resources to users based on their desires and preferences. A hybrid recommendation mechanism based on the singular value decomposition (SVD) method is developed to provide required prediction and recommendation capabilities. Spatial attributes of geospatial resources along with users’ characteristics are utilised to improve the quality of the recommendations. To evaluate the feasibility of the proposed approach, a prototype system was developed and implemented. The results show that the recommender geoportal can relatively predict users’ desires and preferences and provide them with useful geospatial resources.

(Less)
Please use this url to cite or link to this publication:
author
publishing date
type
Contribution to journal
publication status
published
subject
keywords
geo-data, Geoportal, geospatial services, recommender system, singular value decomposition (SVD)
in
Journal of Spatial Science
volume
64
issue
1
pages
23 pages
publisher
Taylor & Francis
external identifiers
  • scopus:85035781298
ISSN
1449-8596
DOI
10.1080/14498596.2017.1397559
language
English
LU publication?
no
id
0c711cdc-5d53-4b57-9c45-bcf96650b973
date added to LUP
2019-03-19 15:46:36
date last changed
2019-11-13 05:28:54
@article{0c711cdc-5d53-4b57-9c45-bcf96650b973,
  abstract     = {<p>Geoportals have been widely implemented by national spatial data infrastructures (NSDI) to provide users with the ability to search for geospatial resources over the Internet. This paper proposes an approach that can improve the functionalities of the conventional geoportals by adding recommendation capabilities to the execution mechanisms of geoportals. The geoportals with recommendation capability are able to analyse users’ activities and recommend geospatial resources to users based on their desires and preferences. A hybrid recommendation mechanism based on the singular value decomposition (SVD) method is developed to provide required prediction and recommendation capabilities. Spatial attributes of geospatial resources along with users’ characteristics are utilised to improve the quality of the recommendations. To evaluate the feasibility of the proposed approach, a prototype system was developed and implemented. The results show that the recommender geoportal can relatively predict users’ desires and preferences and provide them with useful geospatial resources.</p>},
  author       = {Dareshiri, Shokouh and Farnaghi, Mahdi and Sahelgozin, Mohammadreza},
  issn         = {1449-8596},
  keyword      = {geo-data,Geoportal,geospatial services,recommender system,singular value decomposition (SVD)},
  language     = {eng},
  month        = {01},
  number       = {1},
  pages        = {49--71},
  publisher    = {Taylor & Francis},
  series       = {Journal of Spatial Science},
  title        = {A recommender geoportal for geospatial resource discovery and recommendation},
  url          = {http://dx.doi.org/10.1080/14498596.2017.1397559},
  volume       = {64},
  year         = {2019},
}