Deep learning framework for time series classification based on multiple imaging and hybrid quantum neural networks

Time series classification(TSC)has attracted a lot of attention for time series data mining tasks and has been applied in various fields.With the success of deep learning(DL)in computer vision recognition,people are starting to use deep learning to tackle TSC tasks.Quantum neural networks(QNN)have recently demonstrated their superiority over traditional machine learning in methods such as image processing and natural language processing,but research using quantum neural networks to handle TSC tasks has not received enough attention.Therefore,we proposed a learning framework based on multiple imaging and hybrid QNN(MIHQNN)for TSC tasks.We investigate the possibility of converting 1D time series to 2D images and classifying the converted images using hybrid QNN.We explored the differences between MIHQNN based on single time series imaging and MIHQNN based on the fusion of multiple time series imaging.Four quantum circuits were also selected and designed to study the impact of quantum circuits on TSC tasks.We tested our method on several standard datasets and achieved significant results compared to several current TSC methods,demonstrating the effectiveness of MIHQNN.This research highlights the potential of applying quantum computing to TSC and provides the theoretical and experimental background for future research.

Accurate Multi-Scale Feature Fusion CNN for Time Series Classification in Smart Factory

Time series classification(TSC)has attracted various attention in the community of machine learning and data mining and has many successful applications such as fault detection and product identification in the process of building a smart factory.However,it is still challenging for the efficiency and accuracy of classification due to complexity,multi-dimension of time series.This paper presents a new approach for time series classification based on convolutional neural networks(CNN).The proposed method contains three parts:short-time gap feature extraction,multi-scale local feature learning,and global feature learning.In the process of short-time gap feature extraction,large kernel filters are employed to extract the features within the short-time gap from the raw time series.Then,a multi-scale feature extraction technique is applied in the process of multi-scale local feature learning to obtain detailed representations.The global convolution operation with giant stride is to obtain a robust and global feature representation.The comprehension features used for classifying are a fusion of short time gap feature representations,local multi-scale feature representations,and global feature representations.To test the efficiency of the proposed method named multi-scale feature fusion convolutional neural networks(MSFFCNN),we designed,trained MSFFCNN on some public sensors,device,and simulated control time series data sets.The comparative studies indicate our proposed MSFFCNN outperforms other alternatives,and we also provided a detailed analysis of the proposed MSFFCNN.

Self-Supervised Time Series Classification Based on LSTM and Contrastive Transformer

Time series data has attached extensive attention as multi-domain data, but it is difficult to analyze due to its high dimension and few labels. Self-supervised representation learning provides an effective way for processing such data. Considering the frequency domain features of the time series data itself and the contextual feature in the classification task, this paper proposes an unsupervised Long Short-Term Memory(LSTM) and contrastive transformer-based time series representation model using contrastive learning. Firstly, transforming data with frequency domainbased augmentation increases the ability to represent features in the frequency domain. Secondly, the encoder module with three layers of LSTM and convolution maps the augmented data to the latent space and calculates the temporal loss with a contrastive transformer module and contextual loss. Finally, after selfsupervised training, the representation vector of the original data can be got from the pre-trained encoder. Our model achieves satisfied performances on Human Activity Recognition(HAR) and sleepEDF real-life datasets.

Modified Wild Horse Optimization with Deep Learning Enabled Symmetric Human Activity Recognition Model

Traditional indoor human activity recognition(HAR)is a timeseries data classification problem and needs feature extraction.Presently,considerable attention has been given to the domain ofHARdue to the enormous amount of its real-time uses in real-time applications,namely surveillance by authorities,biometric user identification,and health monitoring of older people.The extensive usage of the Internet of Things(IoT)and wearable sensor devices has made the topic of HAR a vital subject in ubiquitous and mobile computing.The more commonly utilized inference and problemsolving technique in the HAR system have recently been deep learning(DL).The study develops aModifiedWild Horse Optimization withDLAided Symmetric Human Activity Recognition(MWHODL-SHAR)model.The major intention of the MWHODL-SHAR model lies in recognition of symmetric activities,namely jogging,walking,standing,sitting,etc.In the presented MWHODL-SHAR technique,the human activities data is pre-processed in various stages to make it compatible for further processing.A convolution neural network with an attention-based long short-term memory(CNNALSTM)model is applied for activity recognition.The MWHO algorithm is utilized as a hyperparameter tuning strategy to improve the detection rate of the CNN-ALSTM algorithm.The experimental validation of the MWHODL-SHAR technique is simulated using a benchmark dataset.An extensive comparison study revealed the betterment of theMWHODL-SHAR technique over other recent approaches.

Attack based on data: a novel perspective to attack sensitive points directly

Adversarial attack for time-series classification model is widely explored and many attack methods are proposed.But there is not a method of attack based on the data itself.In this paper,we innovatively proposed a black-box sparse attack method based on data location.Our method directly attack the sensitive points in the time-series data accord-ing to statistical features extract from the dataset.At frst,we have validated the transferability of sensitive points among DNNs with different structures.Secondly,we use the statistical features extract from the dataset and the sensi-tive rate of each point as the training set to train the predictive model.Then,predicting the sensitive rate of test set by predictive model.Finally,perturbing according to the sensitive rate.The attack is limited by constraining the LO norm to achieve one-point attack.We conduct experiments on several datasets to validate the effectiveness of this method.

Significant wave height prediction through artificial intelligent mode decomposition for wave energy management

The prediction of significant wave height(SWH)is crucial for managing wave energy.While many machine learning studies have focused on accurately predicting SWH values within hours in advance,the primary concern should be given to the level of the wave height for real-world applications.In this paper,a classification framework for the time-series of SWH based on Transformer encoder(TF)and empirical mode decomposition(EMD)is developed,which can provide a lead time of 6 to 48 h with the fixed thresholds of 2 m for high level waves and 1.5 m for low level waves.The performance of this approach is compared to that of three mainstream algorithms with and without EMD features.Results from the datasets collected from buoy measurements in the Atlantic Ocean indicate that the optimal mean accuracy at a lead time of 6 h was 99.1%and the average training time was 75 s,demonstrating the accuracy and efficiency of this proposed model.This study provides valuable tools and references for real-world SWH prediction applications.