A geospatial implementation of a novel delineation clustering algorithm employing the K-means

Tonny Oyana, Kara E. Scott

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The overarching objective of this paper is to introduce a novel Fast, Efficient, and Scalable k-means (FES-k-means) algorithm. This algorithm is designed to increase the overall performance of the standard k-means clustering technique. The FES-k-means algorithm uses a hybrid approach that comprises the k-d tree data structure, the nearest neighbor query, the standard k-means algorithm, and Mashor's adaptation rate. The algorithm is tested using two real datasets and two synthetic datasets and is employed twice on all four datasets. The first trial consisted of previously MIL-SOM trained data, and the second was on raw, untrained data. The approach presented with this method enables unfounded knowledge discovery, otherwise unclaimed by conventional clustering methods. When used in conjunction with the MIL-SOM training technique, the FES-k-means algorithm reduces the computation time and produces quality clusters. In particular, the robust FES-k-means method opens doors to (1) faster cluster production than conventional clustering methods, (2) scalability allowing application in other platforms, and its ability to handle small and large datasets, compact or scattered, and (3) efficient geospatial data analysis of large datasets. All of the above makes FES-k-means live up to defending its well-deserved name-Fast, Efficient, and Scalable k-means (FES-k-means). The findings of this study are vital to the relatively new and expanding subfield of geospatial data management.

Original languageEnglish (US)
Title of host publicationThe European Information Society
Subtitle of host publicationTaking Geoinformation Science One Step Further
Pages135-157
Number of pages23
StatePublished - Dec 1 2008
Event11th AGILE International Conference on Geographic Information Science, AGILE 2008 - Girona, Spain
Duration: May 5 2008May 8 2008

Publication series

NameLecture Notes in Geoinformation and Cartography
ISSN (Print)1863-2351

Other

Other11th AGILE International Conference on Geographic Information Science, AGILE 2008
CountrySpain
CityGirona
Period5/5/085/8/08

Fingerprint

Clustering algorithms
data management
Information management
Data mining
Data structures
Scalability
data analysis
method
ability
management
knowledge
performance

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Geography, Planning and Development
  • Earth-Surface Processes
  • Computers in Earth Sciences

Cite this

Oyana, T., & Scott, K. E. (2008). A geospatial implementation of a novel delineation clustering algorithm employing the K-means. In The European Information Society: Taking Geoinformation Science One Step Further (pp. 135-157). (Lecture Notes in Geoinformation and Cartography).

A geospatial implementation of a novel delineation clustering algorithm employing the K-means. / Oyana, Tonny; Scott, Kara E.

The European Information Society: Taking Geoinformation Science One Step Further. 2008. p. 135-157 (Lecture Notes in Geoinformation and Cartography).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Oyana, T & Scott, KE 2008, A geospatial implementation of a novel delineation clustering algorithm employing the K-means. in The European Information Society: Taking Geoinformation Science One Step Further. Lecture Notes in Geoinformation and Cartography, pp. 135-157, 11th AGILE International Conference on Geographic Information Science, AGILE 2008, Girona, Spain, 5/5/08.
Oyana T, Scott KE. A geospatial implementation of a novel delineation clustering algorithm employing the K-means. In The European Information Society: Taking Geoinformation Science One Step Further. 2008. p. 135-157. (Lecture Notes in Geoinformation and Cartography).
Oyana, Tonny ; Scott, Kara E. / A geospatial implementation of a novel delineation clustering algorithm employing the K-means. The European Information Society: Taking Geoinformation Science One Step Further. 2008. pp. 135-157 (Lecture Notes in Geoinformation and Cartography).
@inproceedings{422985b32b90490f87dc8ab6613c20dc,
title = "A geospatial implementation of a novel delineation clustering algorithm employing the K-means",
abstract = "The overarching objective of this paper is to introduce a novel Fast, Efficient, and Scalable k-means (FES-k-means) algorithm. This algorithm is designed to increase the overall performance of the standard k-means clustering technique. The FES-k-means algorithm uses a hybrid approach that comprises the k-d tree data structure, the nearest neighbor query, the standard k-means algorithm, and Mashor's adaptation rate. The algorithm is tested using two real datasets and two synthetic datasets and is employed twice on all four datasets. The first trial consisted of previously MIL-SOM trained data, and the second was on raw, untrained data. The approach presented with this method enables unfounded knowledge discovery, otherwise unclaimed by conventional clustering methods. When used in conjunction with the MIL-SOM training technique, the FES-k-means algorithm reduces the computation time and produces quality clusters. In particular, the robust FES-k-means method opens doors to (1) faster cluster production than conventional clustering methods, (2) scalability allowing application in other platforms, and its ability to handle small and large datasets, compact or scattered, and (3) efficient geospatial data analysis of large datasets. All of the above makes FES-k-means live up to defending its well-deserved name-Fast, Efficient, and Scalable k-means (FES-k-means). The findings of this study are vital to the relatively new and expanding subfield of geospatial data management.",
author = "Tonny Oyana and Scott, {Kara E.}",
year = "2008",
month = "12",
day = "1",
language = "English (US)",
isbn = "9783540789451",
series = "Lecture Notes in Geoinformation and Cartography",
pages = "135--157",
booktitle = "The European Information Society",

}

TY - GEN

T1 - A geospatial implementation of a novel delineation clustering algorithm employing the K-means

AU - Oyana, Tonny

AU - Scott, Kara E.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - The overarching objective of this paper is to introduce a novel Fast, Efficient, and Scalable k-means (FES-k-means) algorithm. This algorithm is designed to increase the overall performance of the standard k-means clustering technique. The FES-k-means algorithm uses a hybrid approach that comprises the k-d tree data structure, the nearest neighbor query, the standard k-means algorithm, and Mashor's adaptation rate. The algorithm is tested using two real datasets and two synthetic datasets and is employed twice on all four datasets. The first trial consisted of previously MIL-SOM trained data, and the second was on raw, untrained data. The approach presented with this method enables unfounded knowledge discovery, otherwise unclaimed by conventional clustering methods. When used in conjunction with the MIL-SOM training technique, the FES-k-means algorithm reduces the computation time and produces quality clusters. In particular, the robust FES-k-means method opens doors to (1) faster cluster production than conventional clustering methods, (2) scalability allowing application in other platforms, and its ability to handle small and large datasets, compact or scattered, and (3) efficient geospatial data analysis of large datasets. All of the above makes FES-k-means live up to defending its well-deserved name-Fast, Efficient, and Scalable k-means (FES-k-means). The findings of this study are vital to the relatively new and expanding subfield of geospatial data management.

AB - The overarching objective of this paper is to introduce a novel Fast, Efficient, and Scalable k-means (FES-k-means) algorithm. This algorithm is designed to increase the overall performance of the standard k-means clustering technique. The FES-k-means algorithm uses a hybrid approach that comprises the k-d tree data structure, the nearest neighbor query, the standard k-means algorithm, and Mashor's adaptation rate. The algorithm is tested using two real datasets and two synthetic datasets and is employed twice on all four datasets. The first trial consisted of previously MIL-SOM trained data, and the second was on raw, untrained data. The approach presented with this method enables unfounded knowledge discovery, otherwise unclaimed by conventional clustering methods. When used in conjunction with the MIL-SOM training technique, the FES-k-means algorithm reduces the computation time and produces quality clusters. In particular, the robust FES-k-means method opens doors to (1) faster cluster production than conventional clustering methods, (2) scalability allowing application in other platforms, and its ability to handle small and large datasets, compact or scattered, and (3) efficient geospatial data analysis of large datasets. All of the above makes FES-k-means live up to defending its well-deserved name-Fast, Efficient, and Scalable k-means (FES-k-means). The findings of this study are vital to the relatively new and expanding subfield of geospatial data management.

UR - http://www.scopus.com/inward/record.url?scp=84859783957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859783957&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783540789451

T3 - Lecture Notes in Geoinformation and Cartography

SP - 135

EP - 157

BT - The European Information Society

ER -