Bidirectional Long Short-term Memory Network Research Articles

Automatic electrocardiogram detection and classification using bidirectional long short-term memory network improved by Bayesian optimization

Electrocardiogram (ECG) signals contain a significant amount of subtle information that can be used to detect some types of heart dysfunction. The widespread availability of digital ECG and the algorithmic paradigm of the long short-term memory (LSTM) network present an opportunity to substantially improve the accuracy and scalability of automated ECG analysis. However, the number of hidden units and initial learning rate of an LSTM neural network for ECG classification are currently preset based on prior knowledge, which causes the model to reach a sub-optimal state. In this study, an automated ECG detection and classification method using a bidirectional LSTM (BiLSTM) network modified by Bayesian optimization is developed. Bayesian optimization is used to optimize the two hyperparameters of the BiLSTM network: the initial learning rate and the number of hidden layers. By classifying five ECG signals in the MIT-BIH arrhythmia database, the accuracy of the modified network reaches 99.00%, which is 0.86% higher than that before optimization. The results demonstrate that Bayesian optimization can be an effective approach to improving the quality of classifiers based on deep learning. The presented approach can also be considered for generalization to other quasi-periodical biometric signal-based classification tasks in future studies, which may have practical applications.

Long short‐term memory‐based robust and qualitative modal feature identification of non‐stationary low‐frequency oscillation signals in power systems

Low-frequency oscillation (LFO) analysis has become increasingly important in large scale power systems. The current LFO analysis methods regard the measured signal as stationary signal. Some methods, such as Prony and HHT, are too slow for online application. In order to solve these problems, based on the long short-term memory neural network (LSTM), this paper proposes a method to identify LFO modal features rapidly. It is the first time in this field to formulate LFO modal feature classification problem rather than LFO modal identification problem. In order to realize it, first, LFO's frequency and attenuation factor are artificially divided into 12 and 4 segments, respectively. For each segment, more than 20,000 data are generated and tagged as training set and test set. Then two bi-directional LSTM networks are constructed separately, and each network is trained alone with the data of the training set. Finally, the proposed method is tested and verified with the data from the test set, power systems simulation, and phasor measurement unit (PMU) of real power systems. The test results prove effectiveness of the proposed method and its advantage is demonstrated through comparison with the Prony, FFT and EDSNN methods.

A Deep Learning-Based Approach for Generation Expansion Planning Considering Power Plants Lifetime

In Generation Expansion Planning (GEP), the power plants lifetime is one of the most important factors which to the best knowledge of the authors, has not been investigated in the literature. In this article, the power plants lifetime effect on GEP is investigated. In addition, the deep learning-based approaches are widely used for time series forecasting. Therefore, a new version of Long short-term memory (LSTM) networks known as Bi-directional LSTM (BLSTM) networks are used in this paper to forecast annual peak load of the power system. For carbon emissions, the cost of carbon is considered as the penalty of pollution in the objective function. The proposed approach is evaluated by a test network and then applied to Iran power system as a large-scale grid. The simulations by GAMS (General Algebraic Modeling System, Washington, DC, USA) software show that due to consideration of lifetime as a constraint, the total cost of the GEP problem decreases by 5.28% and 7.9% for the test system and Iran power system, respectively.

Analysing behavioural changes in people with dementia using in‐home monitoring technologies

AbstractBackgroundBehavioural changes and neuropsychiatric symptoms such as agitation are common in people with dementia. These symptoms impact the quality of life of people with dementia and can increase the stress on caregivers. This study aims to identify the likelihood of having agitation in people affected by dementia (i.e., patients and carers) using routinely collected data from in‐home monitoring technologies. We have used a digital platform and analytical methods, developed in our previous study, to generate alerts when changes occur in the digital markers collected using in‐home sensing technologies (i.e., vital signs, environmental and activity data). A care monitoring team use the platform and interact with participants and caregivers when an alert is generated.MethodWe have used connected sensory devices to collect environmental markers, including Passive Infra‐Red (PIR), smart power plugs for monitoring home appliance use, motion and door sensors. The environmental marker data have been aggregated within each hour and used to train an agitation risk analysis model. We have trained a model using data collected from 88 homes (∼6 months of data from each home). The proposed model has two components: a self‐supervised transformation learning and an ensemble classification model for agitation likelihood. Ten different neural network encoders are learned to create pseudo‐labels using the samples from the unlabelled data. We use these pseudo‐labels to train a classification model with a convolutional block and a decision layer. The trained convolutional block is then used to learn a latent representation of the data for an ensemble classification block.ResultsComparing with baseline models such as LSTM network, Bidirectional LSTM (BiLSTM) network, VGG, ResNet, Inception, Random Forest (RF), Support Vector Machine (SVM) and Gaussian Process (GP) classifiers, the proposed model performs better in sensitivity (recall) and area under the precision‐recall curve with at most 40% improvement. The recall measure using the 10‐fold cross‐validation technique is 61%.ConclusionThis method can support early interventions and help develop new pathways to support people affected by dementia. A limitation in our current study is that the environmental and movement data is at the home level and not personalised.

Fault detection and diagnosis with a novel source-aware autoencoder and deep residual neural network

The capability of deep learning (DL) techniques for dealing with non-linear, dynamic and correlated data has paved the way for DL-based fault detection and diagnosis (FDD). Among them, autoencoders (AEs) have shown their potential to serve as the fault detection network. However, misclassifying faulty samples that share similar patterns to normal samples is a common drawback of AEs. In this work, a source-aware autoencoder (SAAE) is proposed as an extension of AEs to incorporate faulty samples in the training stage. In SAAE, flexibility in tuning recall and precision trade-off, ability to detect unseen faults and applicability in imbalanced data sets are achieved. Bidirectional long short-term memory (BiLSTM) with skip connections SAAE is designed as the structure of the fault detection network. Further, a deep network with BiLSTM and residual neural network (ResNet) is proposed for the subsequent fault diagnosis step to avoid randomness imposed by the order of the input features. A framework for combining fault detection and fault diagnosis networks is also presented without the assumption of having a perfect fault detection network. A comprehensive comparison among relevant existing techniques in the literature and SAAE-ResNet is also conducted on the Tennessee-Eastman process, which shows the superiority of the proposed FDD method.

A Framework for Segmentation and Classification of Arrhythmia Using Novel Bidirectional LSTM Network

A Framework for Segmentation and Classification of Arrhythmia Using Novel Bidirectional LSTM Network

The Application of Deep Learning Algorithms for PPG Signal Processing and Classification

Photoplethysmography (PPG) is widely used in wearable devices due to its conveniency and cost-effective nature. From this signal, several biomarkers can be collected, such as heart and respiration rate. For the usual acquisition scenarios, PPG is an artefact-ridden signal, which mandates the need for the designated classification algorithms to be able to reduce the noise component effect on the classification. Within the selected classification algorithm, the hyperparameters’ adjustment is of utmost importance. This study aimed to develop a deep learning model for robust PPG wave detection, which includes finding each beat’s temporal limits, from which the peak can be determined. A study database consisting of 1100 records was created from experimental PPG measurements performed in 47 participants. Different deep learning models were implemented to classify the PPG: Long Short-Term Memory (LSTM), Bidirectional LSTM, and Convolutional Neural Network (CNN). The Bidirectional LSTM and the CNN-LSTM were investigated, using the PPG Synchrosqueezed Fourier Transform (SSFT) as the models’ input. Accuracy, precision, recall, and F1-score were evaluated for all models. The CNN-LSTM algorithm, with an SSFT input, was the best performing model with accuracy, precision, and recall of 0.894, 0.923, and 0.914, respectively. This model has shown to be competent in PPG detection and delineation tasks, under noise-corrupted signals, which justifies the use of this innovative approach.

Deep learning reservoir porosity prediction method based on a spatiotemporal convolution bi-directional long short-term memory neural network model

A relationship exists between petrophysical logs and reservoir porosity; however, obtaining analytic solutions is challenging. In this study, we propose a novel method for porosity prediction, which is based on the construction of a new model, namely spatiotemporal convolutional bi-directional long short-term memory (SCBLSTM) network model, based on the convolutional neural network (CNN) and bi-directional long short-term memory (BLSTM) network. The spatial characteristics are extracted by the CNN, and the deep-seated temporal characteristics are extracted by the BLSTM network to ensure that the network model can express the spatiotemporal features of log data. The construction of the model mainly includes network model design, data preprocessing, sample training, and data prediction. After excavating the internal relationship between reservoir porosity and petrophysical logs by training the neural network, reservoir porosity can be predicted directly from petrophysical logs. The simulation verification of the model is performed using the actual logging data. The results show that the SCBLSTM network model can more effectively extract the intrinsic features of logging data and has higher prediction accuracy and stronger stability that other benchmark models. Additionally, the SCBLSTM network model is suitable for the prediction of reservoir physical parameters, which is the problem with multiple series data, thereby providing a new idea for the accurate prediction of reservoir physical parameters.

Simultaneous geometric and thermal error control of gear profile grinder based on analytical correlation between tooth surface error and position error of grinding wheel/workpiece

The gear profile grinder is important to ensure the geometric accuracy of tooth surfaces. However, its machining accuracy is significantly reduced by the geometric and thermal errors. So the geometric and thermal errors should be controlled simultaneously. The geometric error modeling and decoupling of the gear profile grinder are studied, and the mapping relationship between the geometric accuracy of tooth surfaces and machine errors is constructed. To establish a robust thermal error model, a physically-based model is established to prove the feasibility of using the bidirectional long short term memory network (Bi-LSTMN) to establish a self-learning error model. The high-frequency noises are filtered and singular values are reduced by the Savitzky–Golay (SG) filtering algorithm. An attention-gate-based Bi-LSTMN (AGBi-LSTMN) is proposed to reduce the time complexity. Then the mixed-variants whale optimization algorithm (MVWOA) is proposed to optimize the hyper-parameters, and SG-MVWOA-AGBi-LSTMN is proposed for the thermal error prediction. Finally, the simultaneous geometric and thermal error control is implemented. The results show that the total tooth profile precision is increased by ISO level 3, and the tooth profile tilt precision is increased by ISO level 5 with the simultaneous geometric and thermal error control.

Sports Intelligent Assistance System Based on Deep Learning

Traditional sports aid systems analyze sports data via sensors and other types of equipment and can support athletes with retrospective analysis, but they require several sensors and have limited data. This paper examines a sports aid system that uses deep learning to recognize, review, and analyze behaviors through video acquisition and intelligent video sequence processing. This paper’s primary research is as follows: (1) With an eye on the motion assistance system’s application scenarios, the network topology and implementation details of the two-stage Faster R-CNN and the single-stage YOLOv3 target detection algorithms are investigated. Additionally, training procedures are used to enhance the algorithm’s detection performance and training speed. (2) To address the issue of target detection techniques’ low detection performance in complicated backgrounds, an improved scheme from Faster R-CNN is proposed. To begin, a new approach replaces the VGG-16 network in the previous algorithm with a ResNet-101 network. Second, an expansion plan for the dataset is provided. (3) To address the short duration of action video and the high correlation of image sequence data, we present an action recognition method based on LSTM. To begin, we will present a motion decomposition scheme and evaluation index based on the key transaction frame in order to simplify the motion analysis procedure. Second, the spatial features of the frame images are extracted using a convolutional neural network. Besides, the spatial and temporal aspects of the image sequence are fused using a two-layer bidirectional LSTM network. The algorithm suggested in this research has been validated using a golf experiment, and the results are favorable.

An ensemble method to forecast 24-h ahead solar irradiance using wavelet decomposition and BiLSTM deep learning network.

In recent years, the penetration of solar power at residential and utility levels has progressed exponentially. However, due to its stochastic nature, the prediction of solar global horizontal irradiance (GHI) with higher accuracy is a challenging task; but, vital for grid management: planning, scheduling & balancing. Therefore, this paper proposes an ensemble model using the extended scope of wavelet transform (WT) and bidirectional long short term memory (BiLSTM) deep learning network to forecast 24-h ahead solar GHI. The WT decomposes the input time series data into different finite intrinsic model functions (IMF) to extract the statistical features of input time series. Further, the study reduces the number of IMF series by combining the wavelet decomposed components (D1-D6) series on the basis of comprehensive experimental analysis with an aim to improve the forecasting accuracy. Next, the trained standalone BiLSTM networks are allocated to each IMF sub-series to execute the forecasting. Finally, the forecasted values of each sub-series from BiLSTM networks are reconstructed to deliver the final solar GHI forecast. The study performed monthly solar GHI forecasting for one year dataset using one month moving window mechanism for the location of Ahmedabad, Gujarat, India. For the performance comparison, the naïve predictor as a benchmark model, standalone long short term memory (LSTM), gated recurrent unit (GRU), BiLSTM and two other wavelet-based BiLSTM models are also simulated. From the results, it is observed that the proposed model outperforms other models in terms of root mean square error (RMSE) & mean absolute percentage error (MAPE), coefficient of determination (R2) and forecast skill (FS). The proposed model reduces the monthly average RMSE by range from 26.04–58.89%, 5.17–31.35%, 23.26–56.06% & 21.08–57% in comparison with benchmark, standalone BiLSTM, GRU & LSTM networks respectively. On the other hand, the monthly average MAPE is reduced by range from 9 to 51.18%, 12.59–28.14%, 30.43–59.19% & 26.54–58.92% in comparison to benchmark, standalone BiLSTM, GRU & LSTM respectively. Further, the proposed model obtained the value of R2 equal to 0.94 and forecast skill (%) of 47% with reference to the benchmark model.

Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting

Rainfall forecasting has gained utmost research relevance in recent times due to its complexities and persistent applications such as flood forecasting and monitoring of pollutant concentration levels, among others. Existing models use complex statistical models that are often too costly, both computationally and budgetary, or are not applied to downstream applications. Therefore, approaches that use Machine Learning algorithms in conjunction with time-series data are being explored as an alternative to overcome these drawbacks. To this end, this study presents a comparative analysis using simplified rainfall estimation models based on conventional Machine Learning algorithms and Deep Learning architectures that are efficient for these downstream applications. Models based on LSTM, Stacked-LSTM, Bidirectional-LSTM Networks, XGBoost, and an ensemble of Gradient Boosting Regressor, Linear Support Vector Regression, and an Extra-trees Regressor were compared in the task of forecasting hourly rainfall volumes using time-series data. Climate data from 2000 to 2020 from five major cities in the United Kingdom were used. The evaluation metrics of Loss, Root Mean Squared Error, Mean Absolute Error, and Root Mean Squared Logarithmic Error were used to evaluate the models’ performance. Results show that a Bidirectional-LSTM Network can be used as a rainfall forecast model with comparable performance to Stacked-LSTM Networks. Among all the models tested, the Stacked-LSTM Network with two hidden layers and the Bidirectional-LSTM Network performed best. This suggests that models based on LSTM-Networks with fewer hidden layers perform better for this approach; denoting its ability to be applied as an approach for budget-wise rainfall forecast applications.

Fine-grained legal entity annotation: A case study on the Brazilian Supreme Court

The exploration of legal documents in the Brazilian Judiciary context lacks reliable annotated corpus to support the development of new Natural Language Process (NLP) applications. Therefore, this paper presents a step toward exploring legal decisions with Named Entity Recognition (NER) in the Brazilian Supreme Court (STF) context. We aim to present a case study on the fine-grained annotation task of legal decisions, performed by law students as annotators where two levels of nested legal entities were annotated. Nested entities mapped in a preliminary study composed of four coarser legal named entities and twenty-four nested ones (fine-grained). The final result is a corpus of 594 decisions published by the STF annotated by the 76 law students, those with the highest average inter-annotator agreement score. We also present two baselines for NER based on Conditional Random Fields (CRFs) and Bidirectional Long-Short Term Memory Networks (BiLSTMs). This corpus is the first of its kind, the most extensive corpus known in Portuguese dedicated for legal named entity recognition, open and available to better support further research studies in a similar context.

Self-Attention Bi-LSTM Networks for Radar Signal Modulation Recognition

As the electromagnetic environment in battlefields is more and more complex, automatic modulation recognition for radar signals is becoming vital and challenging. Traditional methods are more likely to cause lower recognition accuracy with higher computational complexity in low signal-to-noise ratio (SNR). Feature redundancy especially for handcrafted features is one of the shortcomings of deep-learning-based methods. In this article, a novel end-to-end sequence-based network that consists of a shallow convolutional neural network, a bidirectional long short-term memory (Bi-LSTM) network strengthening with a self-attention mechanism, and a dense neural network is constructed to recognize eight kinds of intrapulse modulations of radar signals. The autocorrelation functions of received radar signals are first calculated as autocorrelation features. Then, these features are employed as inputs of the proposed network which owns significant sequence processing advantages and adaptive selection ability of features. Finally, the proposed network outputs prediction modulations directly. The simulation results verify the robustness and effectiveness of autocorrelation features. And the proposed network achieves about 61.25% accuracy at −20 dB and more than 95% accuracy at −10 dB. Compared with four state-of-the-art networks, the proposed network has better recognition performance especially at low SNRs with much lower computational complexity. Results on measured signals also demonstrate that the proposed network outperforms these four networks.

An ensemble model of convolution and recurrent neural network for skin disease classification

AbstractSkin cancer is one of the rapidly growing diseases in the world. Especially, millions of cases are reported every year by all types of skin cancer in America. Early detection of skin cancer using dermoscopy, the light source, and the magnification device are used to inspect the skin lesions. A dermatologist observed hypodermic structures are normally invisible. However, accurate and effective skin disease classification by humans is not straightforward and requires a long time of practice. Furthermore, it is often inaccurate and difficult to reproduce, being unable to completely use the long‐term dependence connection between specific image key features and image labels even for experienced dermatologists. Therefore, it needs to develop a computer‐aided diagnostic system for reliable skin cancer diagnosis. Classical methods focus on designing and combining hand‐craft features from input data and face vanishing or exploding of loss gradient problem, whereas the bidirectional long short term memory (BLSTM) network does not need any prior knowledge or pre‐designing, and it is an expert in keeping the associated information in both directions. Thus, to improve the classification performance for handling these problems, we proposed a hybrid classification method based on the deep convolutional neural network and stacked BLSTM network. Firstly, deep features are extracted from input skin disease facial images. Next, the sequential features among input data are learned using a dual BLSTM network, where dual BLSTM through max‐pooling, the forward and backward long short term memory (LSTM) hidden states of both the feature matrix and its transpose concatenates for inputting into a dense, fully connected (FC) layer. Finally, the softmax function is used to classify skin disease images. To improve the generalization capability, we evaluate our method on two skin disease image datasets and compare their local image descriptors. The proposed method achieved the best mean accuracy of 91.73%, which shows significant improvements in skin disease classification compared with state‐of‐the‐art skin disease classification methods.

Continuous Human Activity Recognition through Parallelism LSTM with Multi-Frequency Spectrograms

According to the real-living environment, radar-based human activity recognition (HAR) is dedicated to recognizing and classifying a sequence of activities rather than individual activities, thereby drawing more attention in practical applications of security surveillance, health care and human–computer interactions. This paper proposes a parallelism long short-term memory (LSTM) framework with the input of multi-frequency spectrograms to implement continuous HAR. Specifically, frequency-division short-time Fourier transformation (STFT) is performed on the data stream of continuous activities collected by a stepped-frequency continuous-wave (SFCW) radar, generating spectrograms of multiple frequencies which introduce different scattering properties and frequency resolutions. In the designed parallelism LSTM framework, multiple parallel LSTM sub-networks are trained separately to extract different temporal features from the spectrogram of each frequency and produce corresponding classification probabilities. At the decision level, the probabilities of activity classification from these sub-networks are fused by addition as the recognition output. To validate the proposed method, an experimental data set is collected by using an SFCW radar to monitor 11 participants who continuously perform six activities in sequence with three different transitions and random durations. The validation results demonstrate that the average accuracies of the designed parallelism unidirectional LSTM (Uni-LSTM) and bidirectional LSTM (Bi-LSTM) based on five frequency spectrograms are 85.41% and 96.15%, respectively, outperforming traditional Uni-LSTM and Bi-LSTM networks with only a single-frequency spectrogram by 5.35% and 6.33% at least. Additionally, the recognition accuracy of the parallelism LSTM network reveals an upward trend as the number of multi-frequency spectrograms (namely the number of LSTM subnetworks) increases, and tends to be stable when the number reaches 4.

Ocean Observation Data Prediction for Argo Data Quality Control Using Deep Bidirectional LSTM Network

With the rapid development of maritime technologies, a huge amount of ocean data has been acquired through the state-of-the-art ocean equipment to get better understanding and development of ocean. The prediction and correction of oceanic observation data play a fundamental and important role in the oceanic relevant applications, including both civilian and military fields. On the basis of Argo data, aiming at predicting and correcting the oceanic observation data, we propose an ocean temperature and salinity prediction approach in this paper. In our approach, firstly, bounded nonlinear function is utilized for dataset quality control, which can effectively eliminate the influence of spikes or outliers in Argo data. Then, RBF neural network is used for high-resolution Argo dataset construction. Finally, a bidirectional LSTM framework is proposed to predict and analyze the ocean temperature and salinity on the basis of BOA Argo data. Experimental results demonstrate that the proposed bidirectional LSTM framework can accurately predict the ocean temperature and salinity and enable outstanding performance in oceanic observation data prediction and correction. The proposed approach is also important for the realization of Argo dataset automatic quality control.

A Bi-Directional LSTM Network for Estimating Continuous Upper Limb Movement From Surface Electromyography

In human-machine interaction systems, continuous movement estimation methods occupy an important position because they are more natural and intuitive than pattern-recognition methods. Essentially, arm position is decided by the shoulder and elbow joint angles. However, the various deformations of muscles around the shoulder and elbow often lead to difficulties in sensor fixation, which results in a loss of synchronization between the surface electromyography (sEMG) signals and joint angles. In order to accurately estimate movement angles using sEMG in situations where the sEMG is not synchronized with joint angles, we utilized a bi-directional long short-term memory (Bi-LSTM) network rather than other deep learning methods to estimate non-dominant arm movements, based on the sEMG signal from the dominant arm. This estimation protocol was designed to avoid a multiplicity of sensors and to simulate more complicated loss of synchronization problems). The performance of the Bi-LSTM was compared with multilayer perceptrons (MLPs), convolutional neural networks (CNNs), and a long short-term memory network (LSTM). The Pearson correlation coefficient (cc) between the estimated and target joint angle sequences was calculated to evaluate the performance of each neural network. The Wilcoxon signed-rank results showed that the Bi-LSTM model significantly outperformed the MLP, CNN, and LSTM models (tested with completely untrained newly recorded free movements).

Attention-Based STL-BiLSTM Network to Forecast Tourist Arrival

Tourism makes a significant contribution to the economy of almost every country, so accurate demand forecasting can help in better planning for the government and a range of stakeholders involved in the tourism industry and can aid economic sustainability. Machine learning models, and in particular, deep neural networks, can perform better than traditional forecasting models which depend mainly on past observations (e.g., past data) to forecast future tourist arrivals. However, search intensities indices (SII) indicators have recently been included as a forecasting model, which significantly enhances forecasting accuracy. In this study, we propose a bidirectional long short-term memory (BiLSTM) neural network to forecast the arrival of tourists along with SII indicators. The proposed BiLSTM network can remember information from left to right and right to left, which further adds more context for forecasting in memory as compared to a simple long short- term memory (LSTM) network that can remember information only from left to right. A seasonal and trend decomposition using the Loess (STL) approach is utilized to decompose time series tourist arrival data suggested by previous studies. The resultant approach, called STL-BiLSTM, decomposes time series into trend, seasonality, and residual. The trend provides the general direction of the overall data. Seasonality is a regular and predictable pattern which re-occurs at fixed time intervals, and residual is a random fluctuation that is something which cannot be forecast. The proposed BiLSTM network achieves better accuracy than the other methods considered under the current study.

Travel order quantity prediction via attention-based bidirectional LSTM networks

Travel order quantity prediction via attention-based bidirectional LSTM networks