Improving self organizing maps method for data clustering and classification

Zalhan Mohd Zin, Rubiyah Yusof, Ehsan Mesbahi

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

Abstract

The unsupervised learning of Self Organizing Map (SOM) is an effective computational tool in data mining exploration processes. It provides topology preserved data mapping from high-dimensional input space into low-dimensional representation such as two-dimensional map. The visualization and classification of clustered data even with good topological preservation between input and output spaces however are not always easy to be interpreted especially when the data are unknown. The boundaries between the clusters and sub-clusters mapped on SOM map are occasionally not clear. In this paper, we develop an improved SOM (iSOM) method to produce an alternative SOM clustering, classification and visualization. The proposed method firstly groups data into their own classes and then arrange them on the third axis according to their computational distances to winning neurons. The method is demonstrated by computing iSOM clustering and classification on Iris Flowers dataset. The computational results have shown that iSOM was able to provide additional inherent information compared to SOM method. The separation of classes, the positions of a data with respect to other data, the existence of sub-clusters have been clearly presented while classification accuracy has increased.
Original languageEnglish
Title of host publication10th Asian Control Conference (ASCC), 2015
PublisherIEEE
Pages1-7
Number of pages7
ISBN (Electronic)978-1-4799-7862-5
ISBN (Print)978-1-4799-7863-2
DOIs
Publication statusPublished - 10 Sep 2015
Externally publishedYes

Keywords

  • Neurons
  • Training
  • Iris
  • Data visualisation
  • Accuracy
  • Testing
  • Kernel

Cite this

Zin, Z. M., Yusof, R., & Mesbahi, E. (2015). Improving self organizing maps method for data clustering and classification. In 10th Asian Control Conference (ASCC), 2015 (pp. 1-7). IEEE. https://doi.org/10.1109/ASCC.2015.7244588
Zin, Zalhan Mohd ; Yusof, Rubiyah ; Mesbahi, Ehsan. / Improving self organizing maps method for data clustering and classification. 10th Asian Control Conference (ASCC), 2015 . IEEE, 2015. pp. 1-7
@inproceedings{e56c7a0cca8f403d8fa925e85b511104,
title = "Improving self organizing maps method for data clustering and classification",
abstract = "The unsupervised learning of Self Organizing Map (SOM) is an effective computational tool in data mining exploration processes. It provides topology preserved data mapping from high-dimensional input space into low-dimensional representation such as two-dimensional map. The visualization and classification of clustered data even with good topological preservation between input and output spaces however are not always easy to be interpreted especially when the data are unknown. The boundaries between the clusters and sub-clusters mapped on SOM map are occasionally not clear. In this paper, we develop an improved SOM (iSOM) method to produce an alternative SOM clustering, classification and visualization. The proposed method firstly groups data into their own classes and then arrange them on the third axis according to their computational distances to winning neurons. The method is demonstrated by computing iSOM clustering and classification on Iris Flowers dataset. The computational results have shown that iSOM was able to provide additional inherent information compared to SOM method. The separation of classes, the positions of a data with respect to other data, the existence of sub-clusters have been clearly presented while classification accuracy has increased.",
keywords = "Neurons, Training, Iris, Data visualisation, Accuracy, Testing, Kernel",
author = "Zin, {Zalhan Mohd} and Rubiyah Yusof and Ehsan Mesbahi",
year = "2015",
month = "9",
day = "10",
doi = "10.1109/ASCC.2015.7244588",
language = "English",
isbn = "978-1-4799-7863-2",
pages = "1--7",
booktitle = "10th Asian Control Conference (ASCC), 2015",
publisher = "IEEE",
address = "United States",

}

Zin, ZM, Yusof, R & Mesbahi, E 2015, Improving self organizing maps method for data clustering and classification. in 10th Asian Control Conference (ASCC), 2015 . IEEE, pp. 1-7. https://doi.org/10.1109/ASCC.2015.7244588

Improving self organizing maps method for data clustering and classification. / Zin, Zalhan Mohd; Yusof, Rubiyah; Mesbahi, Ehsan.

10th Asian Control Conference (ASCC), 2015 . IEEE, 2015. p. 1-7.

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

TY - GEN

T1 - Improving self organizing maps method for data clustering and classification

AU - Zin, Zalhan Mohd

AU - Yusof, Rubiyah

AU - Mesbahi, Ehsan

PY - 2015/9/10

Y1 - 2015/9/10

N2 - The unsupervised learning of Self Organizing Map (SOM) is an effective computational tool in data mining exploration processes. It provides topology preserved data mapping from high-dimensional input space into low-dimensional representation such as two-dimensional map. The visualization and classification of clustered data even with good topological preservation between input and output spaces however are not always easy to be interpreted especially when the data are unknown. The boundaries between the clusters and sub-clusters mapped on SOM map are occasionally not clear. In this paper, we develop an improved SOM (iSOM) method to produce an alternative SOM clustering, classification and visualization. The proposed method firstly groups data into their own classes and then arrange them on the third axis according to their computational distances to winning neurons. The method is demonstrated by computing iSOM clustering and classification on Iris Flowers dataset. The computational results have shown that iSOM was able to provide additional inherent information compared to SOM method. The separation of classes, the positions of a data with respect to other data, the existence of sub-clusters have been clearly presented while classification accuracy has increased.

AB - The unsupervised learning of Self Organizing Map (SOM) is an effective computational tool in data mining exploration processes. It provides topology preserved data mapping from high-dimensional input space into low-dimensional representation such as two-dimensional map. The visualization and classification of clustered data even with good topological preservation between input and output spaces however are not always easy to be interpreted especially when the data are unknown. The boundaries between the clusters and sub-clusters mapped on SOM map are occasionally not clear. In this paper, we develop an improved SOM (iSOM) method to produce an alternative SOM clustering, classification and visualization. The proposed method firstly groups data into their own classes and then arrange them on the third axis according to their computational distances to winning neurons. The method is demonstrated by computing iSOM clustering and classification on Iris Flowers dataset. The computational results have shown that iSOM was able to provide additional inherent information compared to SOM method. The separation of classes, the positions of a data with respect to other data, the existence of sub-clusters have been clearly presented while classification accuracy has increased.

KW - Neurons

KW - Training

KW - Iris

KW - Data visualisation

KW - Accuracy

KW - Testing

KW - Kernel

U2 - 10.1109/ASCC.2015.7244588

DO - 10.1109/ASCC.2015.7244588

M3 - Conference contribution

SN - 978-1-4799-7863-2

SP - 1

EP - 7

BT - 10th Asian Control Conference (ASCC), 2015

PB - IEEE

ER -

Zin ZM, Yusof R, Mesbahi E. Improving self organizing maps method for data clustering and classification. In 10th Asian Control Conference (ASCC), 2015 . IEEE. 2015. p. 1-7 https://doi.org/10.1109/ASCC.2015.7244588