计算机工程与应用 ›› 2024, Vol. 60 ›› Issue (1): 189-197.DOI: 10.3778/j.issn.1002-8331.2207-0444

• 模式识别与人工智能 • 上一篇    下一篇

增强局部注意力的时间序列分类方法

李克文,柯翠虹,张敏,王晓晖,耿文亮   

  1. 中国石油大学(华东) 计算机科学与技术学院,山东 青岛 266580
  • 出版日期:2024-01-01 发布日期:2024-01-01

Time Series Classification Method with Local Attention Enhancement

LI Kewen, KE Cuihong, ZHANG Min, WANG Xiaohui, GENG Wenliang   

  1. School of Computer Science and Technology, China University of Petroleum (East China), Qingdao, Shandong 266580, China
  • Online:2024-01-01 Published:2024-01-01

摘要: 现有时间序列分类方法普遍基于一种循环网络结构解决时间序列点值耦合问题,无法并行计算,导致计算资源浪费,因此提出一种增强局部注意力的时间序列分类方法。该方法拟合混合距离信息以增加时间序列位置感知能力,将混合距离信息融入自注意矩阵计算中,从而扩展自注意力机制;构造多尺度卷积注意力获取多尺度局部前向信息,以解决标准自注意力机制基于点值计算存在注意力混淆的问题;使用改进后的自注意力机制构造时序自注意分类模块,并行计算处理时间序列分类任务。实验结果表明,与现有时间序列分类方法相比,基于局部注意力增强的时间序列分类方法能够加速收敛,有效提高时序序列分类效果。

关键词: 时间序列分类, 自注意力机制, 位置感知, 多尺度卷积

Abstract: Existing time series classification methods are generally based on a circular network structure to solve the point value coupling problem of time series, which cannot be computed in parallel, resulting in a waste of computing resources. Therefore, this paper proposes a time series classification method with local attention enhancement. The mixed distance information is fitted to increase the position information perception of time series, the mixed distance information is incorporated into the self-attention matrix calculation to expand the self-attention mechanism. Multi-scale convolution attention is constructed to obtain multi-scale local forward information to solve the attention confusion problem in point value calculation of standard self-attention mechanism. The improved self-attention mechanism is used to construct the sequential self-attention classification module, and the time series classification task is processed by parallel computation. The experimental results show that, compared with the existing time series classification methods, the time series classification method based on local attention enhancement can accelerate convergence and effectively improve the classification effect of time series.

Key words: time series classification, self-attention mechanism, position perception, multi-scale convolution