Research on Network Public Opinion Analysis Based on Improved K-means Algorithm

DOI

10.2991/cnci-19.2019.71How to use a DOI?

Keywords

Public opinion analysis, K-means algorithm, improved K-means algorithm, Initial cluster center.

Abstract

In the network public opinion analysis, the K-means algorithm is sensitive to the initial cluster center and outliers. In order to solve this problem, an improved K-means algorithm for optimizing the initial cluster center is proposed. In the improved K-means algorithm, firstly, the outliers of each data object in the dataset are calculated. And the sampling factor  is put forward to obtain the candidate initial center point set. Then, according to the idea of the maximum and minimum distance, k data objects are selected as initial clustering centers from the candidate initial center point set. Finally, a case based on the actual network text data is studied. The case study results show that the new proposed improved K-means algorithm is better than K-means algorithm and K-means++ algorithm in clustering effect. And it is more suitable for network public opinion analysis.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Mingzhong Qi
AU  - Xiongwei Li
PY  - 2019/05
DA  - 2019/05
TI  - Research on Network Public Opinion Analysis Based on Improved K-means Algorithm
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 515
EP  - 520
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.71
DO  - 10.2991/cnci-19.2019.71
ID  - Qi2019/05
ER  -