Proceedings of the 3d International Conference on Applied Social Science Research

K-means and Support Vector Machine in Electric Power Company Benchmarking Management

Authors
Hong-qing Zhang
Corresponding Author
Hong-qing Zhang
Available Online August 2016.
DOI
https://doi.org/10.2991/icassr-15.2016.96How to use a DOI?
Keywords
K-means; SVM; Benchmarking Management; Electric Power Company
Abstract
In the electric power company benchmarking management, implementing classification of the enterprise, the clustering algorithm can set up the model enterprise. It’s very important for the benchmarking management in the electric power company. K-means, as unsupervised learning algorithm, is suitable for processing great sample data, while support vector machine(SVM), as supervised learning algorithm, needs a small number of training samples and is able to obtain the higher classification accuracy. Therefore, the paper presented a classification method based on the combination of SVM and K-means. Using K-means clustered index data first, and then chose some samples which were close to each cluster center as study samples to training SVM classifier and classified all the index data with SVM classifier. Consequently, illustration showed that K-means combined with SVM had higher accuracy than K-means, which testified the validity of it.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
3d International Conference on Applied Social Science Research (ICASSR 2015)
Part of series
Advances in Intelligent Systems Research
Publication Date
August 2016
ISBN
978-94-6252-148-3
ISSN
1951-6851
DOI
https://doi.org/10.2991/icassr-15.2016.96How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Hong-qing Zhang
PY  - 2016/08
DA  - 2016/08
TI  - K-means and Support Vector Machine in Electric Power Company Benchmarking Management
BT  - 3d International Conference on Applied Social Science Research (ICASSR 2015)
PB  - Atlantis Press
SP  - 354
EP  - 357
SN  - 1951-6851
UR  - https://doi.org/10.2991/icassr-15.2016.96
DO  - https://doi.org/10.2991/icassr-15.2016.96
ID  - Zhang2016/08
ER  -