[1]康文轩,陈黎飞,郭躬德.运动序列的时空结构特征表示模型[J].智能系统学报,2023,18(2):240-250.[doi:10.11992/tis.202203011]
KANG Wenxuan,CHEN Lifei,GUO Gongde.Spatiotemporal structure feature representation model for motion sequences[J].CAAI Transactions on Intelligent Systems,2023,18(2):240-250.[doi:10.11992/tis.202203011]
点击复制
KANG Wenxuan,CHEN Lifei,GUO Gongde.Spatiotemporal structure feature representation model for motion sequences[J].CAAI Transactions on Intelligent Systems,2023,18(2):240-250.[doi:10.11992/tis.202203011]
运动序列的时空结构特征表示模型
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
18
期数:
2023年第2期
页码:
240-250
栏目:
学术论文—机器学习
出版日期:
2023-05-05
- Title:
- Spatiotemporal structure feature representation model for motion sequences
- Keywords:
- motion sequence; multidimensional time series; feature extraction; spatiotemporal feature representation model; spatial transformation; key subsequence mining; event sequence; human activity recognition
- 分类号:
- TP301.6
- 摘要:
- 运动序列是一种与运动信号相关的多维时间序列,各个维度序列之间具有高耦合性的特点。现有的多维序列表征方法大多基于维度间相互独立的假设或缺乏可解释性，为此，提出一种适用于运动序列的时空结构特征表示模型及其两阶段构造方法。首先，基于空间变化事件的转换方法，将多维时间序列变换成一维事件序列，以保存序列中的空间结构特性。接着，定义了一种时空结构特征的无监督挖掘算法。基于新定义的表示度度量，该算法从事件序列中提取一组具有代表性的低冗余变长事件元组为时空结构特征。在多个人类行为识别数据集上的实验结果表明，与现有多维时间序列表示方法相比，新模型的特征集更具代表性，在运动序列模式识别领域可以有效提升分类精度。
- Abstract:
- The motion sequence is a multidimensional time series associated with motion signals, and there are high coupling characteristics between different dimensional sequences. Most of the existing multidimensional sequence representation methods are based on the assumption that the dimensions are independent or lack interpretability. As a result, we propose a spatiotemporal structure feature representation model for motion sequences and its two-stage construction method. First, the multidimensional time series are transformed into one-dimensional event sequences using the transformation method of spatially changing events, preserving the spatial structure characteristics of sequences. Second, an unsupervised mining algorithm for spatiotemporal structure features is defined. A collection of representative event grams of various durations and low redundancy are extracted from the event sequence as spatiotemporal structure features using the newly proposed representation measure. The experimental findings on various datasets for recognizing human behavior demonstrate that the new model’s feature set is more representative than the methods currently used to represent multidimensional time series, and it can considerably increase classification accuracy in the domain of motion sequence pattern recognition.
- 参考文献/References:
-
[1] 谭海龙. 多维时间序列的分类技术研究[D]. 杭州: 浙江 大学, 2015: 1?20.
TAN Hailong. Multivariate time series classification technology research[D]. Hanzhou: Zhejiang University, 2015: 1?20.
[2] 任守纲, 张景旭, 顾兴健, 等. 时间序列特征提取方法研究综述[J]. 小型微型计算机系统, 2021, 42(2): 271–278
REN Shougang, ZHANG Jingxu, GU Xingjian. Overview of feature extraction algorithm for time series[J]. Journal of Chinese mini-micro computer systems, 2021, 42(2): 271–278
[3] CORTES C, VAPNIK V, Support-vector networks[J]. Machine learning, 1995, 20: 273?297.
[4] 曹翰林, 唐海娜, 王飞, 等. 轨迹表示学习技术研究进展[J]. 软件学报, 2021, 32(5): 1461–1479
CAO Hanlin, TANG Haina, WANG Fei. Survey on trajectory representation learning techniques[J]. Journal of software, 2021, 32(5): 1461–1479
[5] RABINER L R. A tutorial on hidden Markov models and selected applications in speech recognition[J]. Proceedings of IEEE, 1989, 77: 257–286.
[6] 陈炳鑫, 陈黎飞. 符号序列的预训练HMM分类方法 [J]. 南京大学学报(自然科学版), 2021, 57(1): 52?58.
CHEN Bingxin, CHEN Lifei. A pre-training HMM classification method for symbolic sequences[J]. Journal of Nanjing University (natural science edition). 2021, 57(1): 52?58
[7] HYNDMAN R, KHANDAKAR Y. Automatic time series forecasting: the forecast package for R[J]. Journal of statistical software, 2008, 26(3): 1–22.
[8] YE L, KEOGH E J. Time series shapelets: a novel technique that allows accurate, interpretable and fast classification[J]. Data mining and knowledge discovery, 2010, 22(1-2): 149–182.
[9] XING Z, PEI J, KEOGH E. A brief survey on sequence classification[J]. SIGKDD exploration newsletter, 2010, 12(1): 40–48.
[10] GUO Gongde, CHEN Lifei, YE Yanfang, et al. Cluster validation method for determining the number of clusters in categorical sequences[J]. IEEE transactions on neural networks and learning systems, 2017, 28(12): 2936–2948.
[11] FINK G. Markov models for pattern recognition: from theory to applications[M]. New York: Springer, 2014: 119?201
[12] 钱爱玲. 复杂结构的时间序列数据挖掘与预测方法研究. [D]. 武汉: 华中科技大学, 2011: 13?17.
QIAN Ailing. Research on data mining and forecasting methods over time series with complex structure[D]. Wuhan: Huazhong University, 2011: 13?17.
[13] KASHIPAREKH K, NARWARIYA J, MALHOTRA P, et al. ConvTimeNet: a pre-trained deep convolutional neural network for time series classification[C] //Proc of the 2019 International Joint Conference on Neural Networks. Budapest: IEEE, 2019: 1?8.
[14] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735–80.
[15] ASHISD V, SHASZEER N, PARMAR N, et al. Attention is all you need[C]//Proc of the 31st International Conference on Neural Information Processing Systems. Long beach: Curran Associates Inc., 2017: 5998?6008.
[16] MAKRIDAKIS S, SPILIOTIS E, ASSIMAKOPOULOS V. Statistical and machine learning forecasting methods: Concerns and ways forward[J]. PLoS oNE, 2018, 13(3): e0194889.
[17] BERNDT D, CLIFFORD J. Using dynamic time warping to find patterns in time series[C]//Proc of the 3rd International Conference on Knowledge Discovery and Data Mining. Seattle: AAAI Press, 1994: 359?370.
[18] WANG Jichen, ZHU Weiguo, SUN Yongqi, et al. An effective dynamic spatiotemporal framework with external features information for traffic prediction[J]. Applied Intelligence, 2021, 51: 3159–3173.
[19] SHI Xingjian, CHEN Zhourong, WANG Hao, et al. Convolutional LSTM network: a machine learning approach for precipitation nowcasting[C]//Proc of the 28th Annual Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2015: 802?810.
[20] SUNG W, JUN S, JAEKYUN M. TapNet: neural network augmented with task-adaptive projection for few-shot learning[C]//Proc. of the 36th International Conference on Machine Learning. Long Beach: Curran Associates Inc., 2019, 97: 7115?7123.
[21] XIONG Tengke, WANG Shengrui, JIANG Qingshan, et al. A novel variable-order Markov model for clustering categorical sequences[J]. IEEE transactions on knowledge and data engineering, 2014, 26(10): 2339–2353.
[22] BASU T, MURTHY C. A supervised term selection technique for effective test categorization[J]. International journal of machine learning and cybernetics, 2016, 7: 877–892.
[23] YUAN Liang, WANG Wenjian, CHEN Lifei. Two-stage pruning method for gram-based categorical sequence clustering[J]. International journal machine learning and cybernetics, 2019, 10: 631–640.
[24] DIAS D, MADEO R, ROCHA T, et al. Hand movement recognition for Brazilian Sign Language: a study using distance-based neural networks[C]//Proc of the 2009 International Joint Conference on Neural Networks. Atlanta: IEEE Computer Society, 2009: 697?704.
[25] BEZDEK J. Pattern recognition in handbook of fuzzy computation[M]. Bristol: IOP publishing Ltd., 1998: 1?30.
[26] DAVIDE A, ALESSANDRO G, LUCA O, et al. A public domain dataset for human activity recognition using smartphones[C]//Proc of 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges: Louvain-la-Neuve, 2013: 24?26.
[27] JENNIFER K, GARY W, SAMUEL M. Activity recognition using cell phone accelerometers[J]. ACM SIGKDD explorations newsletter, 2010, 12(2): 74–82.
[28] 程铃纺, 郭躬德, 陈黎飞 符号序列多阶Markov分类 [J]. 计算机应用, 2017, 37(7): 1977?1982
CHENG Lingfang, GUO Gongde, CHEN Lifei. Classification of symbolic sequences with multi-order Markov model[J]. Journal of computer applications. 2017, 37(7): 1977?1982.
备注/Memo
收稿日期:2022-03-06。
基金项目:国家自然科学基金项目（U1805263,61976053）.
作者简介:康文轩,硕士研究生,主要研究方向为序机器学习、数据挖掘;陈黎飞,教授,博士,中国人工智能学会机器学习专业委员会委员,主要研究方向为机器学习、数据挖掘和模式识别。发表学术论文120余篇;郭躬德,教授,博士,中国人工智能学会机器学习专业委员会常务委员,主要研究方向为人工智能、量子计算、机器学习和数据挖掘技术及应用。发表学术论文100余篇
通讯作者:陈黎飞. E-mail：clfei@fjnu.edu.cn
更新日期/Last Update:
1900-01-01