基于矢量量化IFTS的网络流量预测模型
NETWORK TRAFFIC PREDICTION MODEL BASED ON VECTOR QUANTIZATION IFTS
周志强 1杨雪青1
作者信息
- 1. 河南农业职业学院 河南郑州 451450
- 折叠
摘要
针对传统网络流量预测模型存在的局限性,提出一种基于矢量量化直觉模糊时间序列的网络流量预测模型.利用模糊直觉推理有效地表述了网络流量数据中存在的高度模糊性以及不确定性,利用直觉模糊时间序列矢量距离作为评估标准,并且通过坐标平移与质心进行匹配,提升不同时间序列段的分类能力,从而有效地建立网络流量预测模型.通过实验分析可知,提出的预测模型能够提升预测精度并且减少计算复杂度,另外该算法有能力长期预测多个输出.
Abstract
Aimed at the limitation of traditional network traffic prediction model,a new network traffic prediction model based on vector quantization intuitionistic fuzzy time series is proposed.The fuzzy intuitionistic reasoning was used to effectively express the high degree of fuzziness and uncertainty in the network traffic data.The vector distance of intuitionistic fuzzy time series was used as the evaluation standard,and through coordinate translation and centroid matching,the classification ability of different time series was improved,thus effectively establishing network traffic prediction model.According to the experimental analysis,the proposed prediction model can improve the prediction accuracy and reduce the computational complexity.In addition,the algorithm has the ability to predict multiple outputs in a long term.
关键词
直觉模糊时间序列/矢量量化/网络流量/长期预测Key words
Intuitionistic fuzzy time series/Vector quantization/Network traffic/Long-term prediction引用本文复制引用
基金项目
河南省高等学校重点科研项目(19B520030)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据