Time Series Classification Using Multi-Channels Deep Convolutional Neural Networks

By Yi Zheng, Qi Liu, Enhong Chen, Yong Ge, J. Zhao. 2014

640 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Recent advances in convolutional neural networks (2015, methodology): “[27] propose a mixed pooling method which is the combination of max pooling and average pooling.”
Deep Convolutional Neural Networks and Learning ECG Features for Screening Paroxysmal Atrial Fibrillation Patients (2018, methodology): “The first framework is to use deep neural networks as a feature extractor that is combined with discriminative/generative models [18], [19], [21], [24], and the other framework is to use an entire neural network-based classification pipeline [22], [29].”
Deep Learning for Sensor-based Activity Recognition: A Survey (2017, neutral): “(Zheng et al., 2014) Body-worn CNN ADL, heart failure D04, D14”
Exploiting multi-channels deep convolutional neural networks for multivariate time series classification (2015, methodology): “Inspired by the deep feature learning for image classification [11–13], in our preliminary work [14], we introduced a deep learning framework for multivariate time series classification.”
Deep Learning for Sensor-based Human Activity Recognition (2020, methodology): “el" in CNNs, so we use the term "axis" instead in this paper. A commonly used AF strategy is to design a specific neural network for each univariate time series of each sensing channel [176, 191]. The information representations from all channels are concatenated at last for input int”
A Convolutional Neural Network for Fault Classification and Diagnosis in Semiconductor Manufacturing Processes (2016, methodology): “[9] proposed a CNN application in which a multivariate time-series dataset is decomposed into univariate time-series data for each variable, and feature extraction was conducted for each variable in parallel, followed by a classification task with a multilayer perceptron (MLP) with the extracted fea”
Activity recognition in beach volleyball using a Deep Convolutional Neural Network (2017, methodology): “Zheng et al. (2014) used Deep Convolutional Neural Networks (DCNNs) to classify four daily activities from accelerometer data and showed that this approach outperformed a 1-Nearest-Neighbor classifier.”
Deep Convolutional Neural Network Based Regression Approach for Estimation of Remaining Useful Life (2016, methodology): “In [24,26], deep CNN architectures are used and in these architectures all convolutional and pooling filters are one-dimensional which applied along the temporally over individual sensor time series separately.”
Encoding Time Series as Images for Visual Inspection and Classification Using Tiled Convolutional Neural Networks (2014, methodology): “(Zheng et al. 2014) explores supervised feature learning with CNNs to classify multi-channel time series with two datasets.”
Data Augmentation for Time Series Classification using Convolutional Neural Networks (2016, methodology): “This benchmark is composed of relatively small datasets, which has led the authors of [17] not to use them to evaluate their proposed method based on CNN.”
DeepAnT: A Deep Learning Approach for Unsupervised Anomaly Detection in Time Series (2019, methodology): “Generally, CNN and LSTM are used for time series classification problem in literature [7], [30], but we are using CNN (and LSTM for a comparison) for a time series regression problem.”
A Review of Deep Learning Models for Time Series Prediction (2019, background): “such as predicting heart failure from heartbeat data [73] or activity from biometric time series [75].”
Deep Learning for Time-Series Analysis (2017, background): “An architecture that solves this constraint is presented in [60].”
Multivariate LSTM-FCNs for Time Series Classification (2018, methodology): “For this reason, distance based approaches are more successful in classifying multivariate time series data [12].”
TapNet: Multivariate Time Series Classification with Attentional Prototypical Network (2020, methodology): “Recently, deep learning based methods (Karim et al. 2018b)(Zheng et al. 2014) achieve promising performance in time series classiﬁcation task.”