Evaluating Explanation Methods for Multivariate Time Series Classification

被引:2
|
作者
Serramazza, Davide Italo [1 ]
Nguyen, Thu Trang [1 ]
Le Nguyen, Thach [1 ]
Ifrim, Georgiana [1 ]
机构
[1] Univ Coll Dublin, Sch Comp Sci, Dublin, Ireland
基金
爱尔兰科学基金会;
关键词
Time Series Classification; Explanation; Evaluation;
D O I
10.1007/978-3-031-49896-1_11
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Multivariate time series classification is an important computational task arising in applications where data is recorded over time and over multiple channels. For example, a smartwatch can record the acceleration and orientation of a person's motion, and these signals are recorded as multivariate time series. We can classify this data to understand and predict human movement and various properties such as fitness levels. In many applications classification alone is not enough, we often need to classify but also understand what the model learns (e.g., why was a prediction given, based on what information in the data). The main focus of this paper is on analysing and evaluating explanation methods tailored to Multivariate Time Series Classification (MTSC). We focus on saliency-based explanation methods that can point out the most relevant channels and time series points for the classification decision. We analyse two popular and accurate multivariate time series classifiers, ROCKET and dResNet, as well as two popular explanation methods, SHAP and dCAM. We study these methods on 3 synthetic datasets and 2 real-world datasets and provide a quantitative and qualitative analysis of the explanations provided. We find that flattening the multivariate datasets by concatenating the channels works as well as using multivariate classifiers directly and adaptations of SHAP for MTSC work quite well. Additionally, we also find that the popular synthetic datasets we used are not suitable for time series analysis.
引用
收藏
页码:159 / 175
页数:17
相关论文
共 50 条
  • [41] Eigen-entropy based time series signatures to support multivariate time series classification
    Patharkar, Abhidnya
    Huang, Jiajing
    Wu, Teresa
    Forzani, Erica
    Thomas, Leslie
    Lind, Marylaura
    Gades, Naomi
    SCIENTIFIC REPORTS, 2024, 14 (01):
  • [42] Classification of multivariate time series using locality preserving projections
    Weng, Xiaoqing
    Shen, Junyi
    KNOWLEDGE-BASED SYSTEMS, 2008, 21 (07) : 581 - 587
  • [43] Data Augmentation for Multivariate Time Series Classification: An Experimental Study
    Ilbert, Romain
    Hoang, Thai V.
    Zhang, Zonghua
    2024 IEEE 40TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING WORKSHOP, ICDEW, 2024, : 128 - 139
  • [44] Scnet: spectral convolutional networks for multivariate time series classification
    Wu, Xing
    Xing, Xinyu
    Yao, Junfeng
    Qian, Quan
    Song, Jun
    APPLIED INTELLIGENCE, 2025, 55 (06)
  • [45] ST-Tree with interpretability for multivariate time series classification
    Du, Mingsen
    Wei, Yanxuan
    Tang, Yingxia
    Zheng, Xiangwei
    Wei, Shoushui
    Ji, Cun
    NEURAL NETWORKS, 2025, 183
  • [46] Reservoir Computing Approaches for Representation and Classification of Multivariate Time Series
    Bianchi, Filippo Maria
    Scardapane, Simone
    Lokse, Sigurd
    Jenssen, Robert
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 32 (05) : 2169 - 2179
  • [47] Drunk driving detection based on classification of multivariate time series
    Li, Zhenlong
    Jin, Xue
    Zhao, Xiaohua
    JOURNAL OF SAFETY RESEARCH, 2015, 54 : 61 - 67
  • [48] Classification-driven temporal discretization of multivariate time series
    Robert Moskovitch
    Yuval Shahar
    Data Mining and Knowledge Discovery, 2015, 29 : 871 - 913
  • [49] Improving position encoding of transformers for multivariate time series classification
    Navid Mohammadi Foumani
    Chang Wei Tan
    Geoffrey I. Webb
    Mahsa Salehi
    Data Mining and Knowledge Discovery, 2024, 38 : 22 - 48
  • [50] An Examination of the State-of-the-Art for Multivariate Time Series Classification
    Dhariyal, Bhaskar
    Thach Le Nguyen
    Gsponer, Severin
    Ifrim, Georgiana
    20TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING WORKSHOPS (ICDMW 2020), 2020, : 243 - 250