Search

Track smoothness has become an important factor in the safe operation of high-speed trains. In order to ensure the safety of high-speed operations, studies on track smoothness detection methods are constantly improving. This paper presents a track irregularity identification method based on CNN-Bi-LSTM and predicts track irregularity through car body acceleration detection, which is easy to collect and can be obtained by passenger trains, so the model proposed in this paper provides an idea for the development of track irregularity identification method based on conventional vehicles. The first step is construction of the data set required for model training. The model input is the car body acceleration detection sequence, and the output is the irregularity sequence of the same length. The fluctuation trend of the irregularity data is extracted by the HP filtering (Hodrick Prescott Filter) algorithm as the prediction target. The second is a prediction model based on the CNN-Bi-LSTM network, extracting features from the car body acceleration data and realizing the point-by-point prediction of irregularities. Meanwhile, this paper proposes an exponential weighted mean square error with priority inner fitting (EIF-MSE) as the loss function, improving the accuracy of big value data prediction, and reducing the risk of false alarms. In conclusion, the model is verified based on the simulation data and the real data measured by the high-speed railway comprehensive inspection train. [ABSTRACT FROM AUTHOR]

In order to improve the precision of phase recognition and reduce the rate of misdetection, this paper applies the deep learning method to automatic phase recognition. In this paper, an automatic seismic phase recognition model based on the Bi-LSTM network is designed. To test the performance of this model, the STEAD dataset is used for training and testing, and this model is compared with the traditional STA/LTA and AIC methods. The experimental results show that, compared to STA/LTA and AIC methods, the Bi-LSTM network can reduce the misdetection rate by about 8–15%, and improve the RSEM; especially, the prediction error of S-wave is greatly reduced. [ABSTRACT FROM AUTHOR]

In the field of Chinese-named entity recognition, recent research has sparked new interest by combining lexical features with character-based methods. Although this vocabulary enhancement method provides a new perspective, it faces two main challenges: firstly, using character-by-character matching can easily lead to conflicts during the vocabulary matching process. Although existing solutions attempt to alleviate this problem by obtaining semantic information about words, they still lack sufficient temporal sequential or global information acquisition; secondly, due to the limitations of dictionaries, there may be words in a sentence that do not match the dictionary. In this situation, existing vocabulary enhancement methods cannot effectively play a role. To address these issues, this paper proposes a method based on lexical information and spatial features. This method carefully considers the neighborhood and overlap relationships of characters in vocabulary and establishes global bidirectional semantic and temporal sequential information to effectively address the impact of conflicting vocabulary and character fusion on entity segmentation. Secondly, the attention score matrix extracted by the point-by-point convolutional network captures the local spatial relationship between characters without fused vocabulary information and characters with fused vocabulary information, aiming to compensate for information loss and strengthen spatial connections. The comparison results with the baseline model show that the SISF method proposed in this paper improves the F1 metric by 0.72%, 3.12%, 1.07%, and 0.37% on the Resume, Weibo, Ontonotes 4.0, and MSRA datasets, respectively. [ABSTRACT FROM AUTHOR]

In the recent decade, the citation recommendation has emerged as an important research topic due to its need for the huge size of published scientific work. Among other citation recommendation techniques, the widely used content-based filtering (CBF) exploits research articles' textual content to produce recommendations. However, CBF techniques are prone to the well-known cold-start problem. On the other hand, deep learning has shown its effectiveness in understanding the semantics of the text. The present paper proposes a citation recommendation system using deep learning models to classify rhetorical zones of the research articles and compute similarity using rhetorical zone embeddings that overcome the cold-start problem. Rhetorical zones are the predefined linguistic categories having some common characteristics about the text. A deep learning model is trained using ART and CORE datasets with an accuracy of 76 per cent. The final ranked lists of the recommendations have an average of 0.704 normalized discounted cumulative gain (nDCG) score involving ten domain experts. The proposed system is applicable for both local and global context-aware recommendations. [ABSTRACT FROM AUTHOR]

Chinese Folk Drum music is an excellent traditional cultural resource, it has brilliant historical and cultural heritage and excellent traditional cultural connotation. However, the survey found that the social and cultural values, tourism economic values, and national self-confidence embodied in folk drum music, such as Xi'an drum music, are far from being released, and even its own inheritance and development are facing difficulties. The research focuses on the automatic generation of Xi'an drum music, with the aim of further inheriting, developing, and utilizing this exceptional traditional cultural resource. While Artificial Intelligence (AI) music generation has gained popularity in recent years, most platforms primarily focus on modern music rather than Chinese folk music. To address these issues and the unique challenges faced by Xi'an drum music, this paper proposes a Bi-LSTM network-based deep reinforcement learning model. The model incorporates the distinctive characteristics of ancient Chinese music, such as pitch, chord, and mode, and utilizes the Actor-Critic algorithm in reinforcement learning. During the simulation generation stage, an improved method of generating strategies through reward and punishment scores is introduced. Additionally, the model takes into account abstract concept constraints, such as chord progression and music theory rules, which are translated into computer language. By constructing a chord reward mechanism and a music principle reward mechanism, the model achieves harmony constraints and enables the systematic generation of drum music. Experimental results demonstrate that the proposed model, based on Bi-LSTM deep reinforcement learning, can generate Xi'an drum music with high quality and artistic aesthetics. This research contributes to the preservation, development, and utilization of Xi'an drum music, leveraging advancements in AI music generation technology. [ABSTRACT FROM AUTHOR]

Negative emotions of drivers may lead to some dangerous driving behaviors, which in turn lead to serious traffic accidents. However, most of the current studies on driver emotions use a single modality, such as EEG, eye trackers, and driving data. In complex situations, a single modality may not be able to fully consider a driver's complete emotional characteristics and provides poor robustness. In recent years, some studies have used multimodal thinking to monitor single emotions such as driver fatigue and anger, but in actual driving environments, negative emotions such as sadness, anger, fear, and fatigue all have a significant impact on driving safety. However, there are very few research cases using multimodal data to accurately predict drivers' comprehensive emotions. Therefore, based on the multi-modal idea, this paper aims to improve drivers' comprehensive emotion recognition. By combining the three modalities of a driver's voice, facial image, and video sequence, the six classification tasks of drivers' emotions are performed as follows: sadness, anger, fear, fatigue, happiness, and emotional neutrality. In order to accurately identify drivers' negative emotions to improve driving safety, this paper proposes a multi-modal fusion framework based on the CNN + Bi-LSTM + HAM to identify driver emotions. The framework fuses feature vectors of driver audio, facial expressions, and video sequences for comprehensive driver emotion recognition. Experiments have proved the effectiveness of the multi-modal data proposed in this paper for driver emotion recognition, and its recognition accuracy has reached 85.52%. At the same time, the validity of this method is verified by comparing experiments and evaluation indicators such as accuracy and F1 score. [ABSTRACT FROM AUTHOR]

For structural health monitoring (SHM), a complete dataset is crucial for further modal identification analysis and risk warning. Unfortunately, data loss can occur due to sensor failure, transmission system interruption, or hardware failure, which can lead to missing data. Therefore, this study proposes a bidirectional long short-term memory neural network (Bi-LSTM) response recovery method based on variational mode decomposition (VMD) and sparrow search algorithm (SSA) optimization that utilizes the structural response data between multiple sensors and can simultaneously consider temporal and spatial correlations. A dataset containing approximately half a month of monitoring data was collected from a certain project for training, validation, and testing. A publicly available dataset was also referenced to validate the proposed method in this paper. Using the public dataset, under 13 different data loss rates, the VMD + SSA + Bi-LSTM model reduced the RMSE of data reconstruction by an average of 65.01% and 45.35% compared to the Bi-LSTM model and the VMD + Bi-LSTM models, respectively, while the coefficient of determination increased by 62.21% and 11.19%. The data reconstruction method proposed in this paper can accurately reconstruct the variation trends of missing data without the manual optimization of hyperparameters, and the reconstruction results are close to the real data. [ABSTRACT FROM AUTHOR]

Along with the growth of Internet with its numerous potential applications and diverse fields, artificial intelligence (AI) and sentiment analysis (SA) have become significant and popular research areas. Additionally, it was a key technology that contributed to the Fourth Industrial Revolution (IR 4.0). The subset of AI known as emotion recognition systems facilitates communication between IR 4.0 and IR 5.0. Nowadays users of social media, digital marketing, and e-commerce sites are increasing day by day resulting in massive amounts of unstructured data. Medical, marketing, public safety, education, human resources, business, and other industries also use the emotion recognition system widely. Hence it provides a large amount of textual data to extract the emotions from them. The paper presents a systematic literature review of the existing literature published between 2013 to 2023 in text-based emotion detection. This review scrupulously summarized 330 research papers from different conferences, journals, workshops, and dissertations. This paper explores different approaches, methods, different deep learning models, key aspects, description of datasets, evaluation techniques, Future prospects of deep learning, challenges in existing studies and presents limitations and practical implications. [ABSTRACT FROM AUTHOR]

To enhance the software implementation process, developers frequently leverage preexisting code snippets by exploring an extensive codebase. Existing code search tools often rely on keyword- or syntactic-based methods and struggle to fully grasp the semantics and intent behind code snippets. In this paper, we propose a novel hybrid C2B model that combines CodeT5 and bidirectional long short-term memory (Bi-LSTM) for source code search and recommendation. Our proposed C2B hybrid model leverages CodeT5's domain-specific pretraining and Bi-LSTM's contextual understanding to improve code representation and capture sequential dependencies. As a proof-of-concept application, we implemented the proposed C2B hybrid model as a deep neural code search tool and empirically evaluated the model on the large-scale dataset of CodeSearchNet. The experimental findings showcase that our methodology proficiently retrieves pertinent code snippets and surpasses the performance of prior state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Effective air quality prediction models are crucial for the timely prevention and control of air pollution. However, previous models often fail to fully consider air quality's temporal and spatial distribution characteristics. In this study, Xi'an City is used as the study area. Data from 1 January 2019 to 31 October 2020 are used as the training set, while data from 1 November 2020 to 31 December 2020 are used as the test set. This paper proposes a multi-time and multi-site air quality prediction model for Xi'an, leveraging a deep learning network model based on APSO-CNN-Bi-LSTM. The CNN model extracts the spatial features of the input data, the Bi-LSTM model extracts the time series features, and the PSO algorithm with adaptive inertia weight (APSO) optimizes the model's hyperparameters. The results show that the model achieves the best results in terms of MAE and RMSE. Compared to the PSO-SVR, BPTT, CNN-LSTM, and GA-ACO-BP models, the MAE improved by 9.375%, 6.667%, 2.276%, and 4.975%, while the RMSE improved by 8.371%, 8.217%, 6.327%, and 5.293%. These significant improvements highlight the model's accuracy and its promising application prospects. [ABSTRACT FROM AUTHOR]

Recently, interest in the Cyber-Physical System (CPS) has been increasing in the manufacturing industry environment. Various manufacturing intelligence studies are being conducted to enable faster decision-making through various reliable indicators collected from the manufacturing process. Artificial intelligence (AI) and Machine Learning (ML) have advanced enough to give various possibilities of predicting manufacturing time, which can help implement CPS in manufacturing environments, but it is difficult to secure reliability because it is difficult to understand how AI works, and although it can offer good results, it is often not applied to industries. In this paper, Bidirectional Long Short Term Memory (BI-LSTM) is used to predict process execution time, which is an indicator that can be used as a basis for CPS in the manufacturing process, and the Shapley Additive Explanations (SHAP) algorithm is used to explain how artificial intelligence works. The experimental results of this paper, applying manufacturing data, prove that the results derived from SHAP are effective for workers and AI to collaborate. [ABSTRACT FROM AUTHOR]

Recently, the craze of K-POP contents is promoting the development of Korea's cultural and artistic industries. In particular, with the development of various K-POP contents, including dance, as well as the popularity of K-POP online due to the non-face-to-face social phenomenon of the Coronavirus Disease 2019 (COVID-19) era, interest in Korean dance and song has increased. Research on dance Artificial Intelligent (AI), such as artificial intelligence in a virtual environment, deepfake AI that transforms dancers into other people, and creative choreography AI that creates new dances by combining dance and music, is being actively conducted. Recently, the dance creative craze that creates new choreography is in the spotlight. Creative choreography AI technology requires the motions of various dancers to prepare a dance cover. This process causes problems, such as expensive input source datasets and the cost of switching to the target source to be used in the model. There is a problem in that different motions between various dance genres must be considered when converting. To solve this problem, it is necessary to promote creative choreography systems in a new direction while saving costs by enabling creative choreography without the use of expensive motion capture devices and minimizing the manpower of dancers according to consideration of various genres. This paper proposes a system in a virtual environment for automatically generating continuous K-POP creative choreography by deriving postures and gestures based on bidirectional long-short term memory (Bi-LSTM). K-POP dance videos and dance videos are collected in advance as input. Considering a dance video for defining a posture, users who want a choreography, a 3D dance character in the source movie, a new choreography is performed with Bi-LSTM and applied. For learning, considering creativity and popularity at the same time, the next motion is evaluated and selected with probability. If the proposed method is used, the effort for dataset collection can be reduced, and it is possible to provide an intensive AI research environment that generates creative choreography from various existing online dance videos. [ABSTRACT FROM AUTHOR]

The measurement and analysis of vital signs are a subject of significant research interest, particularly for monitoring the driver's physiological state, which is of crucial importance for road safety. Various approaches have been proposed using contact techniques to measure vital signs. However, all of these methods are invasive and cumbersome for the driver. This paper proposes using a non-contact sensor based on continuous wave (CW) radar at 24 GHz to measure vital signs. We associate these measurements with distinct temporal neural networks to analyze the signals to detect and extract heart and respiration rates as well as classify the physiological state of the driver. This approach offers robust performance in estimating the exact values of heart and respiration rates and in classifying the driver's physiological state. It is non-invasive and requires no physical contact with the driver, making it particularly practical and safe. The results presented in this paper, derived from the use of a 1D Convolutional Neural Network (1D-CNN), a Temporal Convolutional Network (TCN), a Recurrent Neural Network particularly the Bidirectional Long Short-Term Memory (Bi-LSTM), and a Convolutional Recurrent Neural Network (CRNN). Among these, the CRNN emerged as the most effective Deep Learning approach for vital signal analysis. [ABSTRACT FROM AUTHOR]

In the last three decades, a lot of work has been done for building Automatic Speech Recognition (ASR) systems for well-established languages such as English, Chinese, etc. However, for implementing a Large Vocabulary Continuous Speech Recognition (LVCSR) system for low resource languages, the research work is also growing rapidly in corpus-focused areas. Hence, there is a requirement of benchmarking large corpus in case of low-resource language like Bangla such that prejudice results can be avoided and limitations can be handled. In the proposed work, an openly available large-scale Bangla speech corpus provided by Google has been used. The work in this paper proposes a combination of image inspired features with well explored cepstral features to build front end feature extraction phase. It uses integration of Convolutional Neural Network (CNN) and bi-directional Long-short Term Memory (bi-LSTM) with Connectionist Temporal Classification (CTC) loss function to implement backend acoustic model. The experiments employ static and dynamic features of Mel-frequency Cepstral Coefficients (MFCC), Constant Q Cepstral Coefficients (CQCC), and Gammatone Cepstral Coefficients (GTCC) techniques, one by one, with Acoustic Ternary Patterns (ATP) features. The proposed work investigates the effect of these various hybrid front-end approaches with CNN, bi-LSTM and integration of these two models. The novelty of this paper lies in the fact that fusion of ATP with cepstral features improves performance of the proposed low resource language ASR system, where the proposed combination of ATP-dynamic CQCC features with integrated backend acoustic model shows a relative improvement of 10–15% in Word Error Rate (WER) over all other experimented combinations. Further to exploit the noise robust nature of GTCC features, the ATP-dynamic GTCC features with integrated CNN-bi-LSTM back-end model are evaluated in noisy scenario, also. [ABSTRACT FROM AUTHOR]

Good prognostic health management (PHM) plays a crucial role in industrial production and other fields. The accurate prediction of remaining useful life (RUL) can ensure good working condition of machines, and the selection of health indicators (HI) is the key to prediction. Due to the length and noise of the data, the selection of features requires a lot of prior knowledge. Therefore, a novel convolution-based attention mechanism bidirectional long and short-term memory (CABLSTM) network is proposed to achieve the end-to-end lifetime prediction of rotating machinery in this paper. Unlike concatenating two networks, the model in this paper is used to convolute the cell states of Bi-LSTM. Firstly, the input signal is performed through CNN to obtain feature information. Secondly, the obtained features are fed into the Bi-LSTM network with attention mechanism for convolution operation to obtain time-frequency information to construct HI. Finally, the training data are normalized to predict RUL. Bi-LSTM can capture features in longer time-frequency information, and attention mechanism can give input influence weight, and highlight its effective characteristics to obtain better prediction accuracy. The complex process of feature extraction, HI construction, and RUL prediction is combined in one algorithm by deep learning. To verify the performance of the method in this paper, experiments were conducted on the bearing dataset of PRONOSTIA, and compared with other methods. The results showed that the method outperforms other methods due to its better accuracy and prediction precision. [ABSTRACT FROM AUTHOR]

Unmanned surface vehicle (USV)'s motion is represented by time-series data that exhibit highly nonlinear and non-stationary features, significantly influenced by environmental factors, such as wind speed and waves, when sailing on the sea. The accurate prediction of USV motion, particularly crucial parameters, such as the roll angle and pitch angle, is imperative for ensuring safe navigation. However, traditional and single prediction models often struggle with low accuracy and fail to capture the intricate spatial–temporal dependencies among multiple input variables. To address these limitations, this paper proposes a prediction approach integrating temporal convolutional network (TCN) and bi-directional long short-term memory network (Bi-LSTM) models, augmented with a temporal pattern attention (TPA) mechanism, termed the TCN-Bi-LSTM-TPA (TBT) USV motion predictor. This hybrid model effectively combines the strengths of TCN and Bi-LSTM architectures to extract long-term temporal features and bi-directional dependencies. The introduction of the TPA mechanism enhances the model's capability to extract spatial information, crucial for understanding the intricate interplay of various motion data. By integrating the features extracted by TCN with the output of the attention mechanism, the model incorporates additional contextual information, thereby improving prediction accuracy. To evaluate the performance of the proposed model, we conducted experiments using real USV motion data and calculated four evaluation metrics: mean square error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and R-squared (R2). The results demonstrate the superior accuracy of the TCN-Bi-LSTM-TPA hybrid model in predicting USV roll angle and pitch angle, validating its effectiveness in addressing the challenges of multivariate USV motion prediction. [ABSTRACT FROM AUTHOR]

Fusion is a state-of-the-art technique to observe the behavioral pattern from time series data. Fusion models efficiently and effectively interpret both linear and nonlinear patterns that are the constraints of an individual model due to feature limitations. In this paper, a three-stage fusion model is proposed to handle time series data and improve stock market forecasting accuracy. In the first phase of fusion, stock market inputs that are constituted with historical data and market sentiments of the targeted stock are pooled along with established technical indicators of the stock market. Market sentiments are examined through sentiment polarity index using big data platform Hadoop. In the second phase, Auto Regressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) are combined to observe linear and nonlinear features of the final stock dataset. In the third phase, an improved Artificial Bee Colony (ABC) algorithm using differential evolution (DE) is examined for the hyperparameter selection of the proposed DE-ABC-Bi-LSTM-ARIMA model for the stock market prediction. In this paper, experiments are performed on established and diversified reported historical datasets Dow Jones Industrial Average index, Nikkei 225 (N225) index, S&P 500 index and NASDAQ GS index. The proposed fusion model DE-ABC-Bi-LSTM-ARIMA outperformed the benchmark models used in this paper. [ABSTRACT FROM AUTHOR]

In recent years, with the rapid development of deep learning, the requirements for the performance of the corresponding real-time recognition system are getting higher and higher. However, the rapid expansion of data volume means that time delay, power consumption, and cost have become problems that cannot be ignored. In this case, the traditional neural network is almost impossible to use to achieve productization. In order to improve the potential problems of a neural network facing a huge number of datasets without affecting the recognition effect, the model compression method has gradually entered people's vision. However, the existing model compression methods still have some shortcomings in some aspects, such as low rank decomposition, transfer/compact convolution filter, knowledge distillation, etc. These problems enable the traditional model compression to cope with the huge amount of computation brought by large datasets to a certain extent, but also make the results unstable on some datasets, and the system performance has not been improved satisfactorily. To address this, we proposed a structured network compression and acceleration method for the convolutional neural network, which integrates the pruned convolutional neural network and the recurrent neural network, and applied it to the lip-recognition system in this paper. [ABSTRACT FROM AUTHOR]

The existing abstractive text summarisation models only consider the word sequence correlations between the source document and the reference summary, and the summary generated by models lacks the cover of the subject of source document due to models' small perspective. In order to make up these disadvantages, a multi‐domain attention pointer (MDA‐Pointer) abstractive summarisation model is proposed in this work. First, the model uses bidirectional long short‐term memory to encode, respectively, the word and sentence sequence of source document for obtaining the semantic representations at word and sentence level. Furthermore, the multi‐domain attention mechanism between the semantic representations and the summary word is established, and the proposed model can generate summary words under the proposed attention mechanism based on the words and sentences. Then, the words are extracted from the vocabulary or the original word sequences through the pointer network to form the summary, and the coverage mechanism is introduced, respectively, into word and sentence level to reduce the redundancy of summary content. Finally, experiment validation is conducted on CNN/Daily Mail dataset. ROUGE evaluation indexes of the model without and with the coverage mechanism are improved respectively, and the results verify the validation of model proposed by this paper. [ABSTRACT FROM AUTHOR]

Long text clustering is of great significance and practical value in data mining, such as information retrieval, text integration, and data compression. Compared with short text clustering, long text clustering involves more semantic information representation and processing, making it a challenging problem. Most recent techniques merely rely on dynamic word embeddings from pre-training as a transfer learning or only based on a simple combination of transformers and traditional clustering methods, which still need to be expanded to short text due to the quadratic computational complexity. In this paper, a novel model combining transfer learning and dynamic feedback called deep embedded clustering with transformer(DEC-transformer) is proposed. To better capture the semantic relationships between sentences in documents, a novel transfer learning technology is firstly applied to long text clustering tasks for pre-training. Unlike previous papers, a two-stage training task is designed by treating semantic representation and text clustering as a united process, and the parameter is dynamically optimized by adaptive feedback to further improve efficiency. Experimental results on the test set show that the proposed model has made great progress in accuracy compared with several benchmarks. Furthermore, it also has good robustness and can get good performance on noisy datasets. [ABSTRACT FROM AUTHOR]

Pyrolysis is one of the thermochemical methods used to produce value-added products from biomass. Thermogravimetric analysis (TGA) is frequently used to examine the energy potential and thermal behavior of biomass, coal, and their blends. The investigation of the TGA data using Artificial Neural Networks (ANN) is one of the most important research areas in recent years. While there are different research papers on the use of Machine Learning (ML) in this field, there is a lack of systematic application of deep learning (DL) algorithms. As such, we applied DL algorithms together with ML algorithms to evaluate the predictive performance of thermal behaviors of proposed bioenergy sources. Thermal behavior of tomato, pepper, eggplant, squash, and cucumber harvest wastes, the equal mass (20%) mixture of them, and the blends of the mixture with coal in the ratios of 20, 33, and 50% under nitrogen atmosphere were investigated by the TGA and ML models. Based on the pyrolysis thermal behavior of the harvest wastes, the eggplant, pepper, tomato, and 5-biomass mixture had the highest conversion potential. According to the thermal behavior of co-pyrolysis of coal and harvest waste mixtures, it had positive effects on pyrolysis conversion degrees and temperature range compared to the coal, and therefore, they can be used as alternative sources for energy production. The MSE and R 2 scores of Bi-directional LSTM demonstrate that an improved performance can be obtained with DL based solutions. Promising results were obtained when the Bi-directional LSTM is applied for modeling the pyrolysis. [ABSTRACT FROM AUTHOR]

Parkinson's Disease (PD) is a brain‐related disease that eventually causes disability and disrupts a person's normal life. Most physicians and researchers try to diagnose it quickly and treat it on time. In recent years, computer science and the field of artificial intelligence (machine learning) have helped researchers find a way to detect the disease early. This article proposes a method that diagnoses Parkinson's disease by analyzing the hand drawing shaped by the individuals using Bi‐Directional Long Short‐Term Memory (Bi‐LSTM) neural network. In addition, this paper proposes a Fuzzy Inferential Classifier for the Dense layer, which classifies the output of LSTM Blocks to the associated classes by modifying the Soft‐max function. Our decision to propose this classifier was because, in some cases, hand‐drawing data related to people with Parkinson's disease have no significant difference with the healthy subjects for the distinguishing and are often very similar to each other. A standard dataset has been used in this paper, which includes spirals drawn test (including static spiral, dynamic spiral, and stability specific point) by a group of healthy people and people with Parkinson's disease. The proposed method has reached 97%, 98.5%, and 100% accuracy rates for the three mentioned spiral tests with a smoother training loss and accuracy plots. In addition, the results outperform state‐of‐the‐art research conducted on this dataset and show at least more than 2.5% improvement in the accuracy rate in comparison. [ABSTRACT FROM AUTHOR]

In this day and age, depression is still one of the biggest problems in the world. If left untreated, it can lead to suicidal thoughts and attempts. There is a need for proper diagnoses of Major Depressive Disorder (MDD) and evaluation of the early stages to stop the side effects. Early detection is critical to identify a variety of serious conditions. In order to provide safe and effective protection to MDD patients, it is crucial to automate diagnoses and make decision-making tools widely available. Although there are various classification systems for the diagnosis of MDD, no reliable, secure method that meets these requirements has been established to date. In this paper, a federated deep learning-based multimodal system for MDD classification using electroencephalography (EEG) and audio datasets is presented while meeting data privacy requirements. The performance of the federated learning (FL) model was tested on independent and identically distributed (IID) and non-IID data. The study began by extracting features from several pre-trained models and ultimately decided to use bidirectional short-term memory (Bi-LSTM) as the base model, as it had the highest validation accuracy of 91% compared to a convolutional neural network and LSTM with 85% and 89% validation accuracy on audio data, respectively. The Bi-LSTM model also achieved a validation accuracy of 98.9% for EEG data. The FL method was then used to perform experiments on IID and non-IID datasets. The FL-based multimodal model achieved an exceptional training and validation accuracy of 99.9% when trained and evaluated on both IID and non-IIID datasets. These results show that the FL multimodal system performs almost as well as the Bi-LSTM multimodal system and emphasize its suitability for processing IID and non-IIID data. Several clients were found to perform better than conventional pre-trained models in a multimodal framework for federated learning using EEG and audio datasets. The proposed framework stands out from other classification techniques for MDD due to its special features, such as multimodality and data privacy for edge machines with limited resources. Due to these additional features, the framework concept is the most suitable alternative approach for the early classification of MDD patients. [ABSTRACT FROM AUTHOR]

When a large car turns at an intersection, it often leads to tragedy because the driver does not pay attention to the incoming car or the dead corner of the line of sight of the car body. On the market, the wheel difference warning system used in large cars generally adds sensors or lenses to confirm whether there are incoming vehicles in the dead corner of the line of sight. However, the accident rate of large vehicles has not been reduced due to the installation of a vision subsidy system. The main reason is that motorcycle and bicycle drivers often neglect to pay attention to the inner wheel difference formed when large vehicles turn, resulting in accidents with large vehicles at intersections. This paper proposes a bidirectional long-term memory neural network for the prediction of the inner wheel path trajectory of large cars, mainly from the perspective of motorcycle riders, through the combination of YOLOv4 and the stacked Bi-LSTM model used in this study to analyze the motion of large cars and predict the inner wheel path trajectory. In this study, the turning trajectory of large vehicles at the intersection is predicted by using an object detection algorithm and cyclic neural network model. Finally, the experiment shows that this study uses the stacked Bi-LSTM trajectory prediction model to predict the next second trajectory with one second trajectory data, and the prediction accuracy is 87.77%; it has an accuracy of 75.75% when predicting the trajectory data of two seconds. In terms of prediction error, the system has a better prediction error than LSTM and Bi-LSTM models. [ABSTRACT FROM AUTHOR]

Safety issues are one of the main limitations for further application of lithium-ion batteries, and battery degradation is an important causative factor. However, current state-of-health (SOH) estimation methods are mostly developed for a single feature and a single operating condition as well as a single battery material system, which consequently makes it difficult to guarantee robustness and generalization. This paper proposes a data-driven and multi-feature collaborative SOH estimation method based on equal voltage interval discharge time, incremental capacity (IC) and differential thermal voltammetry (DTV) analysis for feature extraction. The deep learning model is constructed based on bi-directional long short-term memory (Bi-LSTM) with the addition of attention mechanism (AM) to focus on the important parts of the features. The proposed method is validated based on a NASA dataset and Oxford University dataset, and the results show that the proposed method has high accuracy and strong robustness. The estimated root mean squared error (RMSE) are below 0.7% and 0.3%, respectively. Compared to single features, the collaboration between multiple features and AM resulted in a 25% error improvement, and the capacity rebound is well captured. The proposed method has the potential to be applied online in an end-cloud collaboration system. [ABSTRACT FROM AUTHOR]

Recognition of human activity is a challenging issue, especially in the presence of multiple actions and multiple scenarios. Therefore, in this paper, multi-view multi-modal based human action recognition (HAR) is proposed. Here, initially, motion representation of each image such as Depth motion maps, motion history images, and skeleton images are created from depth, RGB, and skeleton data of RGB-D sensor. After the motion representation, each motion is separately trained by using a 5-stack convolution neural network (5S-CNN). To enhance the recognition rate and accuracy, the skeleton representation is trained using a hybrid 5S-CNN and Bi-LSTM classifier. Then, decision-level fusion is applied to fuse the score value of three motions. Finally, based on the fusion value, the activity of humans is identified. To estimate the efficiency of the suggested 5S-CNN with the Bi-LSTM method, we conduct our experiments using UTD-MHAD. Results show that the suggested HAR method attained better than other existing approaches. [ABSTRACT FROM AUTHOR]

With the development of the urban industry in recent years, air pollution in areas such as factories and streets has become more and more serious. Air quality problems directly affect the normal lives of residents. Effectively predicting the future air condition in the area through relevant historical data has high application value for early warning of this area. Through the study of the previous monitoring data, it is found that the pollutant data of adjacent monitoring stations are correlated in more periods. Therefore, this paper proposes a hybrid model based on CNN and Bi-LSTM, using CNN to synthesize multiple adjacent stations with strong correlations to extract spatial features between data, and using Bi-LSTM to extract features in the time dimension to finally achieve pollutant concentration prediction. Using the historical data of 40 monitoring stations in different locations of Fushun city to conduct research. By comparing with the traditional prediction model, the results prove that the model proposed in this paper has higher accuracy and stronger robustness. [ABSTRACT FROM AUTHOR]

With an increase in the number and types of network attacks, traditional firewalls and data encryption methods can no longer meet the needs of current network security. As a result, intrusion detection systems have been proposed to deal with network threats. The current mainstream intrusion detection algorithms are aided with machine learning but have problems of low detection rates and the need for extensive feature engineering. To address the issue of low detection accuracy, this paper proposes a model for traffic anomaly detection named a deep learning model for network intrusion detection (DLNID), which combines an attention mechanism and the bidirectional long short-term memory (Bi-LSTM) network, first extracting sequence features of data traffic through a convolutional neural network (CNN) network, then reassigning the weights of each channel through the attention mechanism, and finally using Bi-LSTM to learn the network of sequence features. In intrusion detection public data sets, there are serious imbalance data generally. To address data imbalance issues, this paper employs the method of adaptive synthetic sampling (ADASYN) for sample expansion of minority class samples, to eventually form a relatively symmetric dataset, and uses a modified stacked autoencoder for data dimensionality reduction with the objective of enhancing information fusion. DLNID is an end-to-end model, so it does not need to undergo the process of manual feature extraction. After being tested on the public benchmark dataset on network intrusion detection NSL-KDD, experimental results show that the accuracy and F1 score of this model are better than those of other comparison methods, reaching 90.73% and 89.65%, respectively. [ABSTRACT FROM AUTHOR]

The safety of batteries has become a major obstacle to the promotion and application of electric vehicles, and the use of cloud-based vehicle practical big data to summarize the fault knowledge of batteries to improve product quality and reduce maintenance costs has attracted widespread attention from academia and industrial communities. In this paper, a method is proposed to construct the battery fault knowledge graph which supports online knowledge query and fault inference. Reliability models for battery undervoltage, inconsistency, and capacity loss are built based on cloud data, and are deployed and continuously updated in the cloud platform to accommodate the migration of the models to different battery products. A bidirectional long short-term memory (Bi-LSTM) neural network was established for knowledge extraction of fault logs, and the results were imported into Neo4j to form a battery fault knowledge graph. Finally, a fault knowledge online query front-end interface was built to conduct inference tests on battery faults of a manufacturer, which proves the feasibility and effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Daily load forecasting is the basis of the economic and safe operation of a power grid. Accurate prediction results can improve the matching of microgrid energy storage capacity allocation. With the popularization of smart meters, the interaction between residential electricity demand and sources and networks is increasing, and massive data are generated at the same time. Previous forecasting methods suffer from poor targeting and high noise. They cannot make full use of the important information of the load data. This paper proposes a new framework for daily load forecasting of group residents. Firstly, we use the singular value decomposition to address the problem of high dimensions of residential electricity data. Meanwhile, we apply a K-Shape-based group residential load clustering method to obtain the typical residential load data. Secondly, we introduce an empirical mode decomposition method to address the problem of high noise of residential load data. Finally, we propose a Bi-LSTM-Attention model for residential daily load forecasting. This method can make full use of the contextual information and the important information of the daily load of group residents. The experiments conducted on a real data set of a power grid show that our method achieves excellent improvements on five prediction error indicators, such as MAPE, which are significantly smaller than the compared baseline methods. [ABSTRACT FROM AUTHOR]

Radio signals are polluted by noise in the process of channel transmission, which will lead to signal distortion. Noise reduction of radio signals is an effective means to eliminate the impact of noise. Using deep learning (DL) to denoise signals can reduce the dependence on artificial domain knowledge, while traditional signal-processing-based denoising methods often require knowledge of the artificial domain. Aiming at the problem of noise reduction of radio communication signals, a radio communication signal denoising method based on the relativistic average generative adversarial networks (RaGAN) is proposed in this paper. This method combines the bidirectional long short-term memory (Bi-LSTM) model, which is good at processing time-series data with RaGAN, and uses the weighted loss function to construct a noise reduction model suitable for radio communication signals, which realizes the end-to-end denoising of radio signals. The experimental results show that, compared with the existing methods, the proposed algorithm has significantly improved the noise reduction effect. In the case of a low signal-to-noise ratio (SNR), the signal modulation recognition accuracy is improved by about 10% after noise reduction. [ABSTRACT FROM AUTHOR]

In the age of social networks, the number of tweets sent by users has led to a sharp rise in public opinion. Public opinions are closely related to user stances. User stance detection has become an important task in the field of public opinion. However, previous studies have not distinguished between user viewpoints and stances. These studies usually detected stance from the perspective of the tweet level but rarely the user level. Therefore, in this paper, we defined user stance, which is the user viewpoint (support, oppose, and neutral) toward the entire target event process. On this basis, we put forward a user stance detection method based on external commonsense knowledge (such as SenticNet) and environment information (such as a user's historical tweets, topic information, and neighbor tweets) and denote this method as ECKEI. First, in order to better integrate external commonsense knowledge into the neural network, we improved BiLSTM and called it CK-BiLSTM for complementary commonsense information to the memory cell. Secondly, we used LDA to extract the topic of user tweets and designed a topic-driven module to capture the users' neighbors' information. Finally, we used the attention mechanism to integrate information from users' historical tweets and neighbors' tweets obtained through topic information; then, we used the softmax layer to classify user stances into the support, neutral and oppose classes. In this paper, we conducted experiments and assessments on datasets containing information on Brexit and the elections to verify the practicability and effectiveness of our proposed method. Extensive experimental results on the Brexit and elections datasets show that our approach outperforms six baseline methods (SVM-ngram, NB, MTTRE (RNN), Pkudblab (CNN), TAAT, and Aff-feature). We use the average micro-F1 and average accuracy to measure performance on the detection of a user's stance. The ECKEI model makes improvements of 4.30–16.89% and 1.22–16.58% on the Brexit and election datasets, respectively, in terms of average micro-F1. Meanwhile, this model makes improvements of 4.24–17.46% and 0.48–14.64% on the Brexit and election datasets, respectively, in terms of average accuracy. Our model makes improvements of 5.34–17.30% and 2.65–19.73%, respectively, on the Brexit and election datasets in terms of average recall. [ABSTRACT FROM AUTHOR]

Automatic essay scoring (AES) is an essential educational application in natural language processing. This automated process will alleviate the burden by increasing the reliability and consistency of the assessment. With the advances in text embedding libraries and neural network models, AES systems achieved good results in terms of accuracy. However, the actual goals still need to be attained, like embedding essays into vectors with cohesion and coherence, and providing student feedback is still challenging. In this paper, we proposed coherence‐based embedding of an essay into vectors using sentence‐Bidirectional Encoder Representation for Transformers. We trained these vectors on Long short‐term memory and bidirectional long short‐term memory to capture sentence connectivity with other sentences' semantics. We used two datasets: standard ASAP Kaggle and a domain‐specific dataset with almost 2500 responses from 650 students. Our model performed well on both datasets, with an average quadratic weighted kappa score of 0.76. Furthermore, we achieved good results compared to other prescribed models, and we also tested our model on adversarial responses of both datasets and observed decent outcomes. [ABSTRACT FROM AUTHOR]

Outdoor air pollution causes a lot of health problems for humans. Particulate Matter 2.5 (PM2.5), due to its small size, can enter the human respiratory system with ease and cause significant health effects on humans. This makes PM2.5 significant among the various air pollutants. Hence, it is important to measure the value of PM2.5 accurately for better management of air quality. Algorithms for deep learning and machine learning can be used to forecast air quality data. A model that minimizes the prediction error of the PM2.5 forecast is needed. In this paper, a PM2.5 concentration estimation model using Bi-LSTM (Bidirectional Long Short-Term Memory) with meteorological data as predictor variables is proposed. For a better estimation of PM2.5 values, the hyperparameters of the Bi-LSTM model used are tuned using the Osprey Optimization Algorithm (OOA), a recent meta-heuristic algorithm. The model that works with the optimal values of hyperparameters identified by OOA performed better than the other models when they are compared based on evaluation metrics like Mean-Squared Error and R². [ABSTRACT FROM AUTHOR]

Deep learning models have overcome traditional machine learning techniques for text classification domains in the field of natural language processing (NLP). Since, NLP is a branch of machine learning, used for interpreting language, classifying text of interest, and the same can be applied to analyse the medical clinical electronic health records. Medical text consists of lot of rich data which can altogether provide a good insight, by determining patterns from the clinical text data. In this paper, bidirectional‐long short‐term memory (Bi‐LSTM), bi‐LSTM attention and bidirectional encoder representations from transformers (BERT) base models are used to classify the text which are of privacy concern to a person and which should be extracted and can be tagged as sensitive. This text data which we might think not of privacy concern would majorly reveal a lot about the patient's integrity and personal life. Clinical data not only have patient demographic data but lot of hidden data which might go unseen and thus could arise privacy issues. Bi‐LSTM with attention layer is also added on top to realize the importance of critical words which will be of great importance in terms of classification, we are able to achieve accuracy of about 92%. About 206,926 sentences are used out of which 80% are used for training and rest for testing we get accuracy of 90% approx. with Bi‐LSTM alone. The same set of datasets is used for BERT model with accuracy of 93% approx. [ABSTRACT FROM AUTHOR]

This paper introduces a novel diagnostic approach for bearing ball failures: a synergistic implementation of a bidirectional Long Short-Term Memory (LSTM) network, empowered by Gradient Minimum Mean Square. This method leverages deep analysis of operational data from bearings, enabling the precise identification of incipient bearing ball failures at early stages, thus markedly improving prediction accuracy. Our empirical results underscore the superior performance of this composite methodology in accurately detecting a spectrum of five mechanical bearing ball failure types, achieving a substantial enhancement in diagnostic precision. [ABSTRACT FROM AUTHOR]

Mental workload (MW) assessment is crucial for understanding human mental state. Cross-subject MW analysis based on electroencephalogram (EEG) signals is an important way. In this paper, a fine-grained and multi-scale motif (FGMSM) features extraction method is proposed, and the proposed features together with original EEG data are used as the input of bidirectional long short-term memory (Bi-LSTM) to evaluate the cross-subject mental workload. First, the EEG signal of each channel is decomposed based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) algorithm. Second, for the motif structure consisting of three nodes, multi-scale detection is carried out in each intrinsic mode function, and the proportion of each motif structure is extracted as the newly extracted features. Then, the statistical differences of the extracted features between different MW levels are analyzed by using the t -test, and the features with statistical differences are selected for the cross-subject MW assessment. Finally, based on the public dataset with 26 subjects, Bi-LSTM and a variety of machine learning algorithms are used to classify the levels of cross-subject MW. The results show that the Bi-LSTM classification method with the original EEG data and the proposed features show the most positive results. Therefore, the FGMSM features proposed in this paper with Bi-LSTM provide a new technique for the assessment of cross-subject MW based on EEG signals. [ABSTRACT FROM AUTHOR]

A bidirectional long short-term memory (Bi-LSTM) algorithm is proposed to resolve the problem of energy-efficient virtual network embedding. In the VNE process, a large number of attributes can provide information for efficient embedding. This paper divides them into three categories: “network characteristics”, “embedding sequence” and “task type”, and comprehensively analyzes their influence on the embedded performance of virtual networks. This study uses a graph convolutional network (GCN) to extract the network characteristics of virtual and substrate networks. By this approach, we embedded the network-topology graph containing nodes, links, and topological associations onto the input set of the GCN, which rapidly extracts network features. We then used the network features as the model input for the Bi-LSTM neural network to integrate historical and future embedding sequences into the training model. In this process, in conjunction with meta-reinforcement learning to accumulate the experience of various virtual-network tasks, we systematically adjusted model parameters and thus achieved the automatic tuning of neural networks. Simulation results show that when compared to similar existing algorithms. The proposed algorithm improves the acceptance ratio, average ratio of revenue and cost, and reduces the network energy consumption after integrating the network characteristics, embedding sequence, and task type. [ABSTRACT FROM AUTHOR]

The intelligent reflecting surface (IRS) is a two-dimensional (2D) surface with a programmable structure and is composed of many arrays. The arrays are used to supervise electromagnetic wave propagation by altering the electric and magnetic properties of the 2D surface. IRS can influentially convert wireless channels to very effectively enhance spectral efficiency (SE) and communication performance in wireless systems. However, proper channel information is necessary to realize the IRS anticipated gains. The conventional technique has been taken into consideration in recent attempts to fix this issue, which is straightforward but not ideal. A deep learning model which is called the long short-term memory (Bi-LSTM) model can tackle this issue due to its good learning capability and it plays a vital role in enhancing SE. Bi-LSTM can collect data from both forward and backward directions simultaneously to provide improved prediction accuracy. Because of the tremendous benefits of the Bi-LSTM model, in this paper, an IRS-assisted Bi-LSTM model-based multi-user multiple input single output downlink system is proposed for SE improvement. A Wiener filter is used to determine the optimal phase of each IRS element. In the simulation results, the proposed system is compared with other DL models and methods for the SE performance evaluation. The model exhibits satisfactory SE performance with a different signal-to-noise ratio compared to other schemes in the online phase. [ABSTRACT FROM AUTHOR]

Cattle behavior classification technology holds a crucial position within the realm of smart cattle farming. Addressing the requisites of cattle behavior classification in the agricultural sector, this paper presents a novel cattle behavior classification network tailored for intricate environments. This network amalgamates the capabilities of CNN and Bi-LSTM. Initially, a data collection method is devised within an authentic farm setting, followed by the delineation of eight fundamental cattle behaviors. The foundational step involves utilizing VGG16 as the cornerstone of the CNN network, thereby extracting spatial feature vectors from each video data sequence. Subsequently, these features are channeled into a Bi-LSTM classification model, adept at unearthing semantic insights from temporal data in both directions. This process ensures precise recognition and categorization of cattle behaviors. To validate the model's efficacy, ablation experiments, generalization effect assessments, and comparative analyses under consistent experimental conditions are performed. These investigations, involving module replacements within the classification model and comprehensive analysis of ablation experiments, affirm the model's effectiveness. The self-constructed dataset about cattle is subjected to evaluation using cross-entropy loss, assessing the model's generalization efficacy across diverse subjects and viewing perspectives. Classification performance accuracy is quantified through the application of a confusion matrix. Furthermore, a set of comparison experiments is conducted, involving three pertinent deep learning models: MASK-RCNN, CNN-LSTM, and EfficientNet-LSTM. The outcomes of these experiments unequivocally substantiate the superiority of the proposed model. Empirical results underscore the CNN-Bi-LSTM model's commendable performance metrics: achieving 94.3% accuracy, 94.2% precision, and 93.4% recall while navigating challenges such as varying light conditions, occlusions, and environmental influences. The objective of this study is to employ a fusion of CNN and Bi-LSTM to autonomously extract features from multimodal data, thereby addressing the challenge of classifying cattle behaviors within intricate scenes. By surpassing the constraints imposed by conventional methodologies and the analysis of single-sensor data, this approach seeks to enhance the precision and generalizability of cattle behavior classification. The consequential practical, economic, and societal implications for the agricultural sector are of considerable significance. [ABSTRACT FROM AUTHOR]

An efficient tool to execute painless imaging and examine gastrointestinal tract illnesses of the intestine is also known as wireless capsule endoscopy (WCE). Performance, safety, tolerance, and efficacy are the several concerns that make adaptation challenging and wide applicability. In addition, to detect abnormalities, the great importance is the automatic analysis of the WCE dataset. These issues are resolved by numerous vision-based and computer-aided solutions. But, they want further enhancements and do not give the accuracy at the desired level. In order to solve these issues, this paper presents the detection and classification of WCE infected images by a deep neural network and utilizes a bleed image recognizer (BIR) that associates the MobileNetV2 design to classify the images of WCE infected. For the opening-level evaluation, the BIR uses the MobileNetV2 model for its minimum computation power necessity, and then the outcome is sent to the CNN for more processing. Then, Bi-LSTM with an attention mechanism is used to improve the performance level of the model. Hybrid attention Bi-LSTM design yields more accurate classification outcomes. The proposed scheme is implemented in the Python platform and the performance is evaluated by Cohen's kappa, F1-score, recall, accuracy, and precision. The implementation outcomes show that the introduced scheme achieved maximum accuracy of 0.996 with data augmentation with the dataset of WCE images which provided higher outcomes than the others. [ABSTRACT FROM AUTHOR]

Emotions understanding has acquired a significant interest in the last few years because it has introduced remarkable services in many aspects regarding public opinion mining and recognition in the field of marketing, seeking product reviews, reviews of movies, and healthcare issues based on sentiment understanding. This conducted research has utilized the issue of Omicron virus as a case study to implement a emotions analysis framework to explore the global attitude and sentiment toward Omicron variant as an expression of Positive feeling, Neutral, and Negative feeling. Because since December 2021. Omicron variant has gained obvious attention and wide discussions on social media platforms that revealed lots of fears and anxiety feeling, due to its rapid spreading and infection ability between humans that could exceed the Delta variant infection. Therefore, this paper proposes to develop a framework utilizes techniques of natural languages processing (NLP) in deep learning methods using neural network model of Bidirectional-Long-Short-Term-Memory (Bi-LSTM) and deep neural network (DNN) to achieve accurate results. This study utilizes textual data collected and pulled from the Twitter platform (users' tweets) for the time interval from 11-Dec.-2021 to 18-Dec.-2021. Consequently, the overall achieved accuracy for the developed model is 0.946%. The produced results from carrying out the proposed framework for sentiment understanding have recorded Negative sentiment at 42.3%, Positive sentiment at 35.8%, and Neutral sentiment at 21.9% of overall extracted tweets. The acquired accuracy using data of validation for the deployed model is 0.946%. [ABSTRACT FROM AUTHOR]

Nowadays, in a communicating society, fashion is an integral part of a human life, and it is more comfortable and confident when people dress well. Outfit compatibility is not only a combination of different items but also regarding various aspects, such as style, user preferences, and specific occasions. Most of the existing works lead to address the outfit compatibility concerning only style or user preferences, and have no regard for occasions. In this paper, we propose an efficient method for both outfit compatibility and the fill-in-the-blank tasks according to specific occasions. To this end, we utilized an auxiliary classification branch to learn the significantly important features regarding specific occasions. Besides, a sequence to sequence approach is also applied to learn the relationship of different items along with a visual semantic space, which is able to learn the connection between visual features and their semantic presentation. To demonstrate the effectiveness of the proposed method, we conduct experiments on our newly collected Shoplook-Occasion dataset. The experimental results indicate that our proposed method improved the AUC metric from 0.02 to 0.15% and from 0.5 to 4% on accuracy, compared with other approaches for outfit compatibility problem conditioning on specific occasions. [ABSTRACT FROM AUTHOR]

As a hydraulic pump is the power source of a hydraulic system, predicting its remaining useful life (RUL) can effectively improve the operating efficiency of the hydraulic system and reduce the incidence of failure. This paper presents a scheme for predicting the RUL of a hydraulic pump (gear pump) through a combination of a deep convolutional autoencoder (DCAE) and a bidirectional long short-term memory (Bi-LSTM) network. The vibration data were characterized by the DCAE, and a health indicator (HI) was constructed and modeled to determine the degradation state of the gear pump. The DCAE is a typical symmetric neural network, which can effectively extract characteristics from the data by using the symmetry of the encoding network and decoding network. After processing the original vibration data segment, health indicators were entered as a label into the RUL prediction model based on the Bi-LSTM network, and model training was carried out to achieve the RUL prediction of the gear pump. To verify the validity of the methodology, a gear pump accelerated life experiment was carried out, and whole life cycle data were obtained for method validation. The results show that the constructed HI can effectively characterize the degenerative state of the gear pump, and the proposed RUL prediction method can effectively predict the degeneration trend of the gear pump. [ABSTRACT FROM AUTHOR]