《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (1): 90-97.DOI: 10.11772/j.issn.1001-9081.2024010131

• 人工智能 • 上一篇    下一篇

基于信号时态逻辑的深度时序事件检测算法

张思齐1,2, 张金俊1,2, 王天一1,2, 秦小林1,2()   

  1. 1.中国科学院 成都计算机应用研究所,成都 610213
    2.中国科学院大学,北京 100049
  • 收稿日期:2024-02-05 修回日期:2024-03-26 接受日期:2024-03-26 发布日期:2024-05-09 出版日期:2025-01-10
  • 通讯作者: 秦小林
  • 作者简介:张思齐(1995—),男,安徽淮南人,博士研究生,CCF会员,主要研究方向:时态逻辑、时间序列预测、时间序列分析;
    张金俊(2000—),男,江西乐平人,硕士研究生,CCF会员,主要研究方向:图像处理、时间序列预测、自然语言处理;
    王天一(2000—),男,湖北十堰人,硕士研究生,CCF会员,主要研究方向:时间序列预测、自然语言处理;
  • 基金资助:
    国家重点研发计划项目(2023YFB3308601);四川省科技计划项目(2024NSFJQ0035)

Deep temporal event detection algorithm based on signal temporal logic

Siqi ZHANG1,2, Jinjun ZHANG1,2, Tianyi WANG1,2, Xiaolin QIN1,2()   

  1. 1.Chengdu Institute of Computer Application,Chinese Academy of Sciences,Chengdu Sichuan 610213,China
    2.University of Chinese Academy of Sciences,Beijing 100049,China
  • Received:2024-02-05 Revised:2024-03-26 Accepted:2024-03-26 Online:2024-05-09 Published:2025-01-10
  • Contact: Xiaolin QIN
  • About author:ZHANG Siqi, born in 1995, Ph. D. candidate. His research interests include temporal logic, time series forecasting, time series analysis.
    ZHANG Jinjun, born in 2000, M. S. candidate. His research interests include image processing, time series forecasting, natural language processing.
    WANG Tianyi, born in 2000, M. S. candidate. His research interests include time series forecasting, natural language processing.
  • Supported by:
    National Key R&D Program of China(2023YFB3308601);Sichuan Science and Technology Program(2024NSFJQ0035)

摘要:

针对深度事件检测模型对复杂时序事件检测准确性不足和忽略了不同事件间相关性的问题,提出一种基于信号时态逻辑的深度时序事件检测算法DSTL (Deep Signal Temporal Logic)。该算法一方面引入信号时态逻辑框架,并用信号时态逻辑(STL)公式建模时间序列中的事件来综合考虑时间序列上事件的逻辑性和时态性;另一方面采用基于神经网络的基础分类器来检测原子事件的发生情况,并通过STL公式结构和语义来辅助检测复杂事件。另外,使用神经网络模块替代相应的逻辑连接词和时态逻辑算子,从而提供可GPU加速和梯度下降的神经网络模块。通过对6个时间序列数据集的实验,验证了该算法在时序事件检测方面的有效性,并把使用DSTL算法的模型与不使用该算法而使用多层感知机(MLP)、长短期记忆(LSTM)网络和Transformer的深度时间序列分类模型进行比较。实验结果表明,使用DSTL算法的模型在5种事件上的平均F1分数提升了约12%,其中3种跨时间点事件上的平均F1分数提升了约14%,且具备更好的可解释性。

关键词: 时态逻辑, 事件检测, 事件表示, 时间序列, 深度学习

Abstract:

Aiming at the issues of insufficient accuracy in detecting complex temporal events and the neglect of inter-event correlations of deep event detection models, a deep temporal event detection algorithm based on temporal logic, DSTL (Deep Signal Temporal Logic), was proposed. In the algorithm, for one thing, a framework of signal temporal logic was introduced, and events in time series were modeled using Signal Temporal Logic (STL) formulae to consider the logicality and temporality of events in time series comprehensively. For another, a neural network-based base classifier was utilized to detect the occurrence of atomic events, and detection of complex events was aided by structures and semantics of STL formulae. Additionally, neural network modules were employed to replace the corresponding logical conjunctions and temporal logic operators to provide neural network modules supporting GPU acceleration and gradient descent. Through experiments on six time series datasets, the effectiveness of the proposed algorithm in temporal event detection was validated, and the model using DSTL algorithm was compared with deep temporal event detection models using MLP (Multilayer Perceptron), Long Short-Term Memory (LSTM) network and Transformer without using this algorithm. The results indicate that the model using DSTL algorithm has an approximate 12% improvement in average F1 score on five event categories, with an approximate 14% improvement in average F1 score for three categories of cross-time point events, and it has better interpretability.

Key words: temporal logic, event detection, event representation, time series, deep learning

中图分类号: