Using spatial data support for reducing uncertainty in geospatial applications

T. Hong, K. Hart, Leen-Kiat Soh, Ashok K Samal

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Widespread use of GPS devices and ubiquity of remotely sensed geospatial images along with cheap storage devices have resulted in vast amounts of digital data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains now: how to best choose the datasets to solve specific problems? In particular, some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but also lead to erroneous conclusions. On the other hand, excluding some of the datasets hastily might skew the observations drawn. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this paper we use mutual information-a concept popular in information theory as a measure to compute information gain or loss between two datasets-as the basis of computing data support. The flexibility and effectiveness of the approach are demonstrated using an application in the hydrological analysis domain, specifically, watersheds in the state of Nebraska.

Original languageEnglish (US)
Pages (from-to)63-92
Number of pages30
JournalGeoInformatica
Volume18
Issue number1
DOIs
StatePublished - Jan 1 2014

Fingerprint

spatial data
uncertainty
Information use
Information theory
Cost effectiveness
Watersheds
Sensor networks
Global positioning system
Costs
information theory
costs
Uncertainty
cost
data analysis
flexibility
GPS
inclusion
watershed
sensor
analysis

Keywords

  • Data support
  • Mutual information
  • Sensor networks
  • Spatial data mining
  • Time series data

ASJC Scopus subject areas

  • Information Systems
  • Geography, Planning and Development

Cite this

Using spatial data support for reducing uncertainty in geospatial applications. / Hong, T.; Hart, K.; Soh, Leen-Kiat; Samal, Ashok K.

In: GeoInformatica, Vol. 18, No. 1, 01.01.2014, p. 63-92.

Research output: Contribution to journalArticle

@article{32e8f9faddfc41ebbb3c1bc8ecde43db,
title = "Using spatial data support for reducing uncertainty in geospatial applications",
abstract = "Widespread use of GPS devices and ubiquity of remotely sensed geospatial images along with cheap storage devices have resulted in vast amounts of digital data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains now: how to best choose the datasets to solve specific problems? In particular, some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but also lead to erroneous conclusions. On the other hand, excluding some of the datasets hastily might skew the observations drawn. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this paper we use mutual information-a concept popular in information theory as a measure to compute information gain or loss between two datasets-as the basis of computing data support. The flexibility and effectiveness of the approach are demonstrated using an application in the hydrological analysis domain, specifically, watersheds in the state of Nebraska.",
keywords = "Data support, Mutual information, Sensor networks, Spatial data mining, Time series data",
author = "T. Hong and K. Hart and Leen-Kiat Soh and Samal, {Ashok K}",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s10707-013-0177-z",
language = "English (US)",
volume = "18",
pages = "63--92",
journal = "GeoInformatica",
issn = "1384-6175",
publisher = "Kluwer Academic Publishers",
number = "1",

}

TY - JOUR

T1 - Using spatial data support for reducing uncertainty in geospatial applications

AU - Hong, T.

AU - Hart, K.

AU - Soh, Leen-Kiat

AU - Samal, Ashok K

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Widespread use of GPS devices and ubiquity of remotely sensed geospatial images along with cheap storage devices have resulted in vast amounts of digital data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains now: how to best choose the datasets to solve specific problems? In particular, some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but also lead to erroneous conclusions. On the other hand, excluding some of the datasets hastily might skew the observations drawn. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this paper we use mutual information-a concept popular in information theory as a measure to compute information gain or loss between two datasets-as the basis of computing data support. The flexibility and effectiveness of the approach are demonstrated using an application in the hydrological analysis domain, specifically, watersheds in the state of Nebraska.

AB - Widespread use of GPS devices and ubiquity of remotely sensed geospatial images along with cheap storage devices have resulted in vast amounts of digital data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains now: how to best choose the datasets to solve specific problems? In particular, some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but also lead to erroneous conclusions. On the other hand, excluding some of the datasets hastily might skew the observations drawn. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this paper we use mutual information-a concept popular in information theory as a measure to compute information gain or loss between two datasets-as the basis of computing data support. The flexibility and effectiveness of the approach are demonstrated using an application in the hydrological analysis domain, specifically, watersheds in the state of Nebraska.

KW - Data support

KW - Mutual information

KW - Sensor networks

KW - Spatial data mining

KW - Time series data

UR - http://www.scopus.com/inward/record.url?scp=84893776842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893776842&partnerID=8YFLogxK

U2 - 10.1007/s10707-013-0177-z

DO - 10.1007/s10707-013-0177-z

M3 - Article

AN - SCOPUS:84893776842

VL - 18

SP - 63

EP - 92

JO - GeoInformatica

JF - GeoInformatica

SN - 1384-6175

IS - 1

ER -