Abstract
Data mining plays a critical role in sustainable decision-making. Although the k-prototypes algorithm is one of the best-known algorithms for clustering both numeric and categorical data, clustering a large number of spatial objects with mixed numeric and categorical attributes is still inefficient due to complexity. In this paper, we propose an efficient grid-based k-prototypes algorithm, GK-prototypes, which achieves high performance for clustering spatial objects. The first proposed algorithm utilizes both maximum and minimum distance between cluster centers and a cell, which can reduce unnecessary distance calculation. The second proposed algorithm as an extension of the first proposed algorithm, utilizes spatial dependence; spatial data tends to be similar to objects that are close. Each cell has a bitmap index which stores the categorical values of all objects within the same cell for each attribute. This bitmap index can improve performance if the categorical data is skewed. Experimental results show that the proposed algorithms can achieve better performance than the existing pruning techniques of the k-prototypes algorithm.
| Original language | English |
|---|---|
| Article number | 2614 |
| Journal | Sustainability (Switzerland) |
| Volume | 10 |
| Issue number | 8 |
| DOIs | |
| Publication status | Published - 2018 Jul 25 |
Fingerprint
Keywords
- Clustering
- Data mining
- Grid-based k-prototypes
- Spatial data
- Sustainability
ASJC Scopus subject areas
- Geography, Planning and Development
- Renewable Energy, Sustainability and the Environment
- Management, Monitoring, Policy and Law
Cite this
An efficient grid-based k-prototypes algorithm for sustainable decision-making on spatial objects. / Jang, Hong Jun; Kim, Byoungwook; Kim, Jongwan; Jung, Soon Young.
In: Sustainability (Switzerland), Vol. 10, No. 8, 2614, 25.07.2018.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - An efficient grid-based k-prototypes algorithm for sustainable decision-making on spatial objects
AU - Jang, Hong Jun
AU - Kim, Byoungwook
AU - Kim, Jongwan
AU - Jung, Soon Young
PY - 2018/7/25
Y1 - 2018/7/25
N2 - Data mining plays a critical role in sustainable decision-making. Although the k-prototypes algorithm is one of the best-known algorithms for clustering both numeric and categorical data, clustering a large number of spatial objects with mixed numeric and categorical attributes is still inefficient due to complexity. In this paper, we propose an efficient grid-based k-prototypes algorithm, GK-prototypes, which achieves high performance for clustering spatial objects. The first proposed algorithm utilizes both maximum and minimum distance between cluster centers and a cell, which can reduce unnecessary distance calculation. The second proposed algorithm as an extension of the first proposed algorithm, utilizes spatial dependence; spatial data tends to be similar to objects that are close. Each cell has a bitmap index which stores the categorical values of all objects within the same cell for each attribute. This bitmap index can improve performance if the categorical data is skewed. Experimental results show that the proposed algorithms can achieve better performance than the existing pruning techniques of the k-prototypes algorithm.
AB - Data mining plays a critical role in sustainable decision-making. Although the k-prototypes algorithm is one of the best-known algorithms for clustering both numeric and categorical data, clustering a large number of spatial objects with mixed numeric and categorical attributes is still inefficient due to complexity. In this paper, we propose an efficient grid-based k-prototypes algorithm, GK-prototypes, which achieves high performance for clustering spatial objects. The first proposed algorithm utilizes both maximum and minimum distance between cluster centers and a cell, which can reduce unnecessary distance calculation. The second proposed algorithm as an extension of the first proposed algorithm, utilizes spatial dependence; spatial data tends to be similar to objects that are close. Each cell has a bitmap index which stores the categorical values of all objects within the same cell for each attribute. This bitmap index can improve performance if the categorical data is skewed. Experimental results show that the proposed algorithms can achieve better performance than the existing pruning techniques of the k-prototypes algorithm.
KW - Clustering
KW - Data mining
KW - Grid-based k-prototypes
KW - Spatial data
KW - Sustainability
UR - http://www.scopus.com/inward/record.url?scp=85050472794&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85050472794&partnerID=8YFLogxK
U2 - 10.3390/su10082614
DO - 10.3390/su10082614
M3 - Article
AN - SCOPUS:85050472794
VL - 10
JO - Sustainability
JF - Sustainability
SN - 2071-1050
IS - 8
M1 - 2614
ER -