Spatial Coordinate Trial : Converting Non-Spatial Data Dimension for DBSCAN

Arriyanti, Eka; Arfyanti, Ita; Adytia, Pitrasacha

doi:10.23919/eecsi48112.2019.8977130

Cited by 1 publication

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because, despite it is a spatial clustering based on Euclidean distance, the data type is numeric and it can detect clusters of arbitrary shape [5]. This is what has been attested by [6] with Spatial Coordinate Way; SCW, the non-spatial data converting way, where clustering of DBSCAN considers the data to be clustered as points (X, Y) in integers (not geospatial coordinates) and with Density Based Spatial Clustering Application with Noise-radius parameter range; DBSCAN-rpr. This is for finding the most formed clusters with the smallest noise, and all of formed clusters are proven good.…”

Section: Introductionmentioning

confidence: 99%

A Synthesis of Optimal Unknown Number Clustering System and Categorical

2020

IJATCSE

View full text Add to dashboard Cite

Big data makes data dynamic and research moves from sample to population. These circumstances imply data mining, specifically clustering, should be able to produce the formed clusters of population as much as possible, thus leaving only a little "junk" of data. Unknown number clustering data mining is a term for set of clustering algorithms which cluster without defining number of formed clusters. Without intending to let others and their data types, this research proposes a synthesis of optimal unknown number clustering system and categorical, because the excellence of DBSCAN algorithm performance on spatial and non-spatial data, and the expectance of standard clustering system presence for population statistics. After reanalyzing its Spatial Coordinate Way and the radius parameter range, the synthesis is done with structural analysis based on general data mining steps for framing the complete clustering process. Then, the result of analysis is compared against the cluster analysis requirements to convince the conclusion on the system synthesis. The categorical data without weighting are also possibly converting into spatial form, but imply qualitative value. Therefore, the synthesis recommends that the standard clustering system should have special items on unweighted categorical data in addition to a numeric type.

show abstract

Section: Introductionmentioning

confidence: 99%

A Synthesis of Optimal Unknown Number Clustering System and Categorical

2020

IJATCSE

View full text Add to dashboard Cite

show abstract

Spatial Coordinate Trial : Converting Non-Spatial Data Dimension for DBSCAN

Cited by 1 publication

References 7 publications

A Synthesis of Optimal Unknown Number Clustering System and Categorical

A Synthesis of Optimal Unknown Number Clustering System and Categorical

Contact Info

Product

Resources

About