Search

Software defects are a critical issue in software development that can lead to system failures and cause significant financial losses. Predicting software defects is a vital aspect of ensuring software quality. This can significantly impact both saving time and reducing the overall cost of software testing. During the software defect prediction (SDP) process, automated tools attempt to predict defects in the source codes based on software metrics. Several SDP models have been proposed to identify and prevent defects before they occur. In recent years, recurrent neural network (RNN) techniques have gained attention for their ability to handle sequential data and learn complex patterns. Still, these techniques are not always suitable for predicting software defects due to the problem of imbalanced data. To deal with this problem, this study aims to combine a bidirectional long short-term memory (Bi-LSTM) network with oversampling techniques. To establish the effectiveness and efficiency of the proposed model, the experiments have been conducted on benchmark datasets obtained from the PROMISE repository. The experimental results have been compared and evaluated in terms of accuracy, precision, recall, f-measure, Matthew's correlation coefficient (MCC), the area under the ROC curve (AUC), the area under the precision-recall curve (AUCPR) and mean square error (MSE). The average accuracy of the proposed model on the original and balanced datasets (using random oversampling and SMOTE) was 88%, 94%, And 92%, respectively. The results showed that the proposed Bi-LSTM on the balanced datasets (using random oversampling and SMOTE) improves the average accuracy by 6 and 4% compared to the original datasets. The average F-measure of the proposed model on the original and balanced datasets (using random oversampling and SMOTE) were 51%, 94%, And 92%, respectively. The results showed that the proposed Bi-LSTM on the balanced datasets (using random oversampling and SMOTE) improves the average F-measure by 43 and 41% compared to the original datasets. The experimental results demonstrated that combining the Bi-LSTM network with oversampling techniques positively affects defect prediction performance in datasets with imbalanced class distributions. [ABSTRACT FROM AUTHOR]

The paper developing a resilient speech classification system for individuals with voice disorders poses a formidable challenge due to the significant variability and distortions inherent in vocal signals. This article outlines the steps to create an effective classification system for pathological speech. The first step involved applying speech enhancement processing using the minimum mean square error (MMSE) enhancer to improve voice input data quality and intelligibility. Secondly, a multi-stream approach combined various acoustic vectors based on human auditory perception, including mel-spectrogram images, mel frequency cepstral coefficients (MFCC), power normalized cepstral coefficients (PNCC), and prosodic parameters like F0, Jitter, and Shimmer. Finally, a deep machine learning incorporating both a convolutional neural network (CNN) and a bidirectional long short-term memory (BiLSTM) network was employed to process these enhanced characteristics in a multi-stream framework, resulting in a powerful classification system architecture. In our experiments, we utilized two subsets from the Massachusetts Eye and Ear Infirmary (MEEI) database, each involving distinct causes of voice disorders. The first subset consisted of voice recordings from patients with vocal nodules, paralysis, and polyps, while the second subset included recordings from patients with mild ventricular compression, A–P squeezing, and gastric reflux. The results we obtained reveal that the CNN-BiLSTM system, coupled with a robust speech analysis interface based on the multi-stream approach and enhanced by the minimum mean square error (MMSE) processing, achieved the highest accuracy rates. [ABSTRACT FROM AUTHOR]

Accurate short-term load forecasting (STLF) plays an essential role in sustainable energy development. Specifically, energy companies can efficiently plan and manage their generation capacity, lessening resource wastage and promoting the overall efficiency of power resource utilization. However, existing models cannot accurately capture the nonlinear features of electricity data, leading to a decline in the forecasting performance. To relieve this issue, this paper designs an innovative load forecasting method, named Prophet–CEEMDAN–ARBiLSTM, which consists of Prophet, Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), and the residual Bidirectional Long Short-Term Memory (BiLSTM) network. Specifically, this paper firstly employs the Prophet method to learn cyclic and trend features from input data, aiming to discern the influence of these features on the short-term electricity load. Then, the paper adopts CEEMDAN to decompose the residual series and yield components with distinct modalities. In the end, this paper designs the advanced residual BiLSTM (ARBiLSTM) block as the input of the above extracted features to obtain the forecasting results. By conducting multiple experiments on the New England public dataset, it demonstrates that the Prophet–CEEMDAN–ARBiLSTM method can achieve better performance compared with the existing Prophet-based ones. [ABSTRACT FROM AUTHOR]

This paper addresses the critical need for accurate photovoltaic (PV) power generation predictions to ensure efficient grid integration and management, especially considering the variability and intermittency of solar power. By exploring advanced deep learning techniques, including Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and a hybrid CNN-RNN model, the study aims to enhance the accuracy and reliability of solar power forecasts. The CNN model achieved an accuracy of 0.84, while the RNN reached 0.94, with the highest accuracy of 0.99 attained by the hybrid CNN-RNN model. These models provide vital tools for mitigating fluctuations in solar power output, improving grid stability, and optimizing energy distribution. The study contributes to the advancement of renewable energy forecasting, helping to ensure a more sustainable and reliable energy future, while also supporting efforts to reduce CO2 emissions and combat climate change.

The inherent volatility in electricity prices exerts a significant impact on the dynamic nature of the electricity market, shaping the decision-making processes of its stakeholders. Precise Electricity Price Forecasting (EPF) plays a pivotal role in enabling energy suppliers to optimize their bidding strategies, mitigate transactional risks, and capitalize on market opportunities, thereby ensuring alignment with the true economic value of energy transactions. Hence, this study proposes an advanced deep learning model for forecasting electricity prices one day in ahead. The model leverages the synergistic capabilities of Convolutional Neural Networks (CNN) and bidirectional Long Short-Term Memory networks (BiLSTM), operating concurrently with an autoregressive (AR) component, denoted as CNN-BiLSTM-AR. The integration of the AR model alongside CNN-BiLSTM enhances overall performance by exploiting AR’s proficiency in capturing transient linear dependencies. Simultaneously, CNN-BiLSTM excels in assimilating spatial and protracted temporal features. Moreover, the research delves into the implications of incorporating hyperparameter optimization (HPO) techniques, such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Random Search (RS). The effectiveness of the model is evaluated using two distinct European datasets sourced from the UK and German electricity markets. Performance metrics, including Root Mean Square Error (RMSE) and Mean Absolute Error (MAE), serve as benchmarks for assessment. Finally, the findings underscore the notable performance enhancement achieved through the implementation of HPO methods in conjunction with the proposed model. Especially, the PSO-CNN-BiLSTM-AR model demonstrates substantial reductions in RMSE and MAE, amounting to 16.7% and 23.46%, respectively, for the German electricity market.

Cardiovascular disease such as ischemic heart disease and stroke are the most dangerous diseases in the WHO stats. Myocardial Infarction (MI), an ischemic disease of the heart, occurs due to a sudden blockage in the coronary arteries that supply blood to the heart causing a lack of oxygen and nutrients. The MI patient needs continuous monitoring using electrocardiography, the latter is always at risk of developing complications such as arrhythmias. As a solution, we proposed an internet of things (IoT) based ECG system for monitoring, the application layer was reserved for the detection of MI and arrhythmias using artificial intelligence so that the patients can keep being monitored even outside health facilities. For this purpose, this paper proposed a hybrid Convolutional Neural Network (CNN) – Bidirectional Long Short-Term Memory (BiLSTM) approach to classify ECG signals and evaluates its performance by using raw and preprocessed data, and comparing the results to related studies. Two datasets have been used in this classification. The results were promising, the model has scored 99.00% accuracy on raw data classifying 4 classes, and 99.73% accuracy on a larger preprocessed data for 3 classes classification. The proposed model is suitable to serve in our monitoring task.

Detecting fake speech in voice-based authentication systems is crucial for reliability. Traditional methods often struggle because they can't handle the complex patterns over time. Our study introduces an advanced approach using deep learning, specifically Long Short-Term Memory (LSTM) and Bidirectional LSTM (BiLSTM) models, tailored for identifying fake speech based on its temporal characteristics. We use speech signals with cepstral features like Mel-frequency cepstral coefficients (MFCC), Constant Q cepstral coefficients (CQCC), and open-source Speech and Music Interpretation by Large-space Extraction (OpenSMILE) to directly learn these patterns. Testing on the ASVspoof 2019 Logical Access dataset, we focus on metrics such as min-tDCF, Equal Error Rate (EER), Recall, Precision, and F1-score. Our results show that LSTM and BiLSTM models significantly enhance the reliability of spoof speech detection systems.

A novel approach for generative time series simulation of electricity price scenarios is presented. A “Time Series Simulation Conditional Generative Adversarial Network” (TSS-CGAN) generates short-term electricity price scenarios. In particular, the network is capable of generating a 24-dimensional output vector that corresponds to the expected behavior of electricity markets. The model can replace typical approaches from financial mathematics like statistical factor models to model the price distribution around a given forecast. The data cover a 3-year period from 2020 to 2023. Our empirical study is conducted on the EPEX SPOT market in Europe. An electricity price scenario includes the prices of the hourly contracts of a day-ahead auction at the EPEX SPOT power exchange. The model uses multivariate time series as input factors, consisting of point forecasts of electricity prices and fundamental data on generation and load profiles. The architecture of a TSS-CGAN is based on the idea of Conditional Generative Adversarial Networks combined with 1D Convolutional Neural Networks and Bidirectional Long Short-Term Memory. The model is evaluated using qualitative and quantitative criteria. For the evaluation, 10,000 simulations of a test period are carried out. Qualitative criteria are whether the model follows certain electricity market-specific regularities and depicts them adequately. The quantitative analysis includes common error metric, compared to benchmark models, like DeepAR, Prophet and Temporal Fusion Transformer, the examination of the quantile ranges, the error distribution and a sensitivity analysis. The results show that the TSS-CGAN outperforms benchmark models such as DeepAR by reducing the continuous ranked probability score by 50% and considers market-specific circumstances such as the production of fluctuating energies and reacts correctly to changes in the corresponding variables.

Introduction: Eye movement is one of the cues used in human--machine interface technologies for predicting the intention of users. The developing application in eye movement event detection is the creation of assistive technologies for paralyzed patients. However, developing an effective classifier is one of the main issues in eye movement event detection. Methods: In this paper, bidirectional long short-term memory (BILSTM) is proposed along with hyperparameter tuning for achieving effective eye movement event classification. The Lévy flight and interactive crossoverbased reptile search algorithm (LICRSA) is used for optimizing the hyperparameters of BILSTM. The issues related to overfitting are avoided by using fuzzy data augmentation (FDA), and a deep neural network, namely, VGG- 19, is used for extracting features from eye movements. Therefore, the optimization of hyperparameters using LICRSA enhances the classification of eye movement events using BILSTM. Results and Discussion: The proposed BILSTM--LICRSA is evaluated by using accuracy, precision, sensitivity, F1-score, area under the receiver operating characteristic (AUROC) curve measure, and area under the precision--recall curve (AUPRC) measure for four datasets, namely, Lund2013, collected dataset, GazeBaseR, and UTMultiView. The gazeNet, human manual classification (HMC), and multi-source information-embedded approach (MSIEA) are used for comparison with the BILSTM--LICRSA. The F1-score of BILSTM--LICRSA for the GazeBaseR dataset is 98.99%, which is higher than that of the MSIEA. [ABSTRACT FROM AUTHOR]

In the modern era of technology, monitoring and controlling abnormal human activity is essentially required as these activities may harm society through physical harm to a human being, or by spreading hate crimes on the World Wide Web. Although many authors have contributed to address this problem, a desired solution that may work in a real-time scenario has yet to be achieved. Recently, deep learning models have gained attraction as processing power for a large volume of data. However, there is little work based on deep learning models for detecting abnormal human activity classification that has been done till now. In the proposed framework, a deep-learning method has been used to detect abnormal human activity by combining a convolutional neural network (CNN), a Recurrent Neural Network (RNN), and an attention module for attending the specific spatiotemporal characteristics from unprocessed video streams. This proposed architecture can accurately classify an aberrant human activity with its special category after processing the video. The proposed architecture's analytical results show an accuracy of 96.94%, 98.95%, and 62.04% with UCF50, UCF110, and UCF crime datasets, which is compared with the results of state-of-the-art algorithms (SOTA). [ABSTRACT FROM AUTHOR]

The fetal electrocardiogram (FECG) records changes in the graph of fetal cardiac action potential during conduction, reflecting the developmental status of the fetus in utero and its physiological cardiac activity. Morphological alterations in the FECG can indicate intrauterine hypoxia, fetal distress, and neonatal asphyxia early on, enhancing maternal and fetal safety through prompt clinical intervention, thereby reducing neonatal morbidity and mortality. To reconstruct FECG signals with clear morphological information, this paper proposes a novel deep learning model, CBLS-CycleGAN. The model's generator combines spatial features extracted by the CNN with temporal features extracted by the BiLSTM network, thus ensuring that the reconstructed signals possess combined features with spatial and temporal dependencies. The model's discriminator utilizes PatchGAN, employing small segments of the signal as discriminative inputs to concentrate the training process on capturing signal details. Evaluating the model using two real FECG signal databases, namely "Abdominal and Direct Fetal ECG Database" and "Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeat Annotations", resulted in a mean MSE and MAE of 0.019 and 0.006, respectively. It detects the FQRS compound wave with a sensitivity, positive predictive value, and F1 of 99.51%, 99.57%, and 99.54%, respectively. This paper's model effectively preserves the morphological information of FECG signals, capturing not only the FQRS compound wave but also the fetal P-wave, T-wave, P-R interval, and ST segment information, providing clinicians with crucial diagnostic insights and a scientific foundation for developing rational treatment protocols. [ABSTRACT FROM AUTHOR]

Cross-project defect prediction (CPDP) aims to predict software defects in a target project domain by leveraging information from different source project domains, allowing testers to identify defective modules quickly. However, CPDP models often underperform due to different data distributions between source and target domains, class imbalances, and the presence of noisy and irrelevant instances in both source and target projects. Additionally, standard features often fail to capture sufficient semantic and contextual information from the source project, leading to poor prediction performance in the target project. To address these challenges, this research proposes Smote Correlation and Attention Gated recurrent unit based Long Short-Term Memory optimization (SCAG-LSTM), which first employs a novel hybrid technique that extends the synthetic minority over-sampling technique (SMOTE) with edited nearest neighbors (ENN) to rebalance class distributions and mitigate the issues caused by noisy and irrelevant instances in both source and target domains. Furthermore, correlation-based feature selection (CFS) with best-first search (BFS) is utilized to identify and select the most important features, aiming to reduce the differences in data distribution among projects. Additionally, SCAG-LSTM integrates bidirectional gated recurrent unit (Bi-GRU) and bidirectional long short-term memory (Bi-LSTM) networks to enhance the effectiveness of the long short-term memory (LSTM) model. These components efficiently capture semantic and contextual information as well as dependencies within the data, leading to more accurate predictions. Moreover, an attention mechanism is incorporated into the model to focus on key features, further improving prediction performance. Experiments are conducted on apache_lucene, equinox, eclipse_jdt_core, eclipse_pde_ui, and mylyn (AEEEM) and predictor models in software engineering (PROMISE) datasets and compared with active learning-based method (ALTRA), multi-source-based cross-project defect prediction method (MSCPDP), the two-phase feature importance amplification method (TFIA) on AEEEM and the two-phase transfer learning method (TPTL), domain adaptive kernel twin support vector machines method (DA-KTSVMO), and generative adversarial long-short term memory neural networks method (GB-CPDP) on PROMISE datasets. The results demonstrate that the proposed SCAG-LSTM model enhances the baseline models by 33.03%, 29.15% and 1.48% in terms of F1-measure and by 16.32%, 34.41% and 3.59% in terms of Area Under the Curve (AUC) on the AEEEM dataset, while on the PROMISE dataset it enhances the baseline models' F1-measure by 42.60%, 32.00% and 25.10% and AUC by 34.90%, 27.80% and 12.96%. These findings suggest that the proposed model exhibits strong predictive performance. [ABSTRACT FROM AUTHOR]

One of the significant application of robot vision technique that serves in biometric verification systems is Face Recognition (FR) technology. However, the effectiveness of the model is affected by disturbance from the real-world environment, including alterations in lighting, facial occlusion, and fluctuation in poses. Even though the recognition of faces has gained popularity due to the variety of applications, it still remains a complicated task to recognize because of an enormous variety of biometric data features. In the past few years, FR technology has seen a lot of development in this area of biometric and computer vision-oriented applications. The most important steps in creating an accurate FR system are the extraction of features and the categorization of these extracted features. The conventional method of feature extraction includes frequency domain features or the Eigenface technique. However, they are not robust to altering external factors like occlusion, illumination, and posture. Using deep learning techniques, a unique approach to address this problem in FR is suggested to recognize faces successfully. The images are aggregated in the initial phase. It is then uploaded to the Viola–Jones Face Detector tools to find the face. The detected face images are then subjected to the Adaptive Deep Convolutional Generative Adversarial Network (ADCGAN) to remove the occlusion from the facial images, where the parameters are subsequently optimized using the Modified Random Variable-based Galactic Swarm Optimizations (MRV-GSO) algorithm. The occlusion-removed face images are then given to the Multi-Cascaded Attentive and Adaptive Deep Learning Network (MCADN) model, where the outputs from the Dilated DenseNet and Residual Network (ResNet) are serially passed to the Bidirectional Long-Short Term Memory (Bi-LSTM) model forming the MCDAN. The MRV-GSO algorithm is performed to tune the hyperparameters in the MCADN model to produce the final recognized facial images. As a result of conducting several experimental studies, it is proved that the generated model outperformed standard approaches in terms of effective recognition rate. [ABSTRACT FROM AUTHOR]

Product reviews are becoming a more popular tool for businesses and individuals when making judgements about purchases. Spammers create synthesized reviews to either promote certain items or denigrate those of rivals to make money. As a result, in recent years, both the business and research sectors have paid close attention to the detection of false opinion spam. Customers' decision-making is severely harmed by false opinion spam in service or product evaluations. It's becoming difficult to identify false opinion spam. Accordingly, the article proposed to detect deceptive opinion spam based on a hybrid deep learning technique. Initially, the model was tested using deceptive reviews gathered from several online forums. To identify deceptive reviews, many researchers at the moment create models based on a single text attribute. On the contrary, deceptive reviewers will decisively copy the wording style of legitimate evaluations while submitting reviews. These text-feature-based techniques may or may not be successful. As a result, the research suggested an ensemble multiple-feature selection technique of the Extra tree classifier to extract information based on a variety of features, including text, behaviour, and deceptive scoring features. In addition, a data resampling approach is used that integrates the Borderline-SMOTE algorithm to reduce the effects of the high dimensional imbalanced class category distribution. For detecting deceptive reviews, the article developed a hybrid technique of Bidirectional Long Short-Term Memory (Bi-LSTM) with a Capsule Neural Network to detect the positive and negative false opinions spam. The model optimizes the dynamic routing algorithm and changes the structure of the conventional capsule network without sacrificing classification performance, leading to high model accuracy. The model performance is evaluated using Python software. The study assesses the suggested model using data from two distinct domains (hotel and restaurant) as a standard benchmark. The experimental results demonstrate the advantage of neural models with higher accuracy of 99% respectively, showing that the suggested neural model greatly outperforms the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Accurately predicting stock prices is crucial for reducing investment-related risks in decision-making. Contemporary challenges to financial behavior, posed by environmental issues such as pollution and climate change, have received limited attention in existing studies on capital market predictability. This paper focuses on the Shanghai Stock Exchange Composite Index (SSEC) and employs air quality and weather data from the Shanghai area as input variables. Subsequently, a hybrid prediction model is constructed by integrating Grey Relational Analysis (GRA), Wavelet Decomposition (WD), and Bidirectional Long Short-Term Memory (BiLSTM) neural networks. The objective is to achieve precise predictions of closing prices. Additionally, this study validates the feasibility of incorporating environmental factors as input variables for stock price prediction, using the Shenzhen Component Index (SZI) and Hang Seng Index (HSI) as case studies, while also assessing the applicability of the GRA-WD-BiLSTM model. The findings demonstrate that the GRA-WD-BiLSTM model exhibits superior applicability and prediction performance in stock price forecasting, with respective prediction accuracies of 95.93%, 93.02%, and 97.07% when accounting for environmental factors. The incorporation of GRA and WD contributes to enhancing single models' performance while integrating air quality and weather factors, which prove valuable in accurately predicting stock prices. The findings also indicate that the impact of regional environmental factors on local stock exchange index prices shows variability. [ABSTRACT FROM AUTHOR]

Thrust constitutes a pivotal performance parameter for aircraft engines. Thrust, being an indispensable parameter in control systems, has garnered significant attention, prompting numerous scholars to propose various methods and algorithms for its estimation. However, research methods for estimating the thrust of the micro-turbojet engines used in unmanned aerial vehicles are relatively scarce. Therefore, this paper proposes a thrust estimator for micro-turbojet engines based on DBO (dung beetle optimization) utilizing bidirectional long short-term memory (BiLSTM) and a convolutional neural network (CNN). Furthermore, the efficacy of the proposed model is further validated through comparative analysis with others in this paper. [ABSTRACT FROM AUTHOR]

Speech production is a complex sequential process which involve the coordination of various articulatory features. Among them tongue being a highly versatile active articulator responsible for shaping airflow to produce targeted speech sounds that are intellectual, clear, and distinct. This paper presents a novel approach for predicting tongue and lip articulatory features involved in a given speech acoustics using a stacked Bidirectional Long Short-Term Memory (BiLSTM) architecture, combined with a one-dimensional Convolutional Neural Network (CNN) for post-processing with fixed weights initialization. The proposed network is trained with two datasets consisting of simultaneously recorded speech and Electromagnetic Articulography (EMA) datasets, each introducing variations in terms of geographical origin, linguistic characteristics, phonetic diversity, and recording equipment. The performance of the model is assessed in Speaker Dependent (SD), Speaker Independent (SI), corpus dependent (CD) and cross corpus (CC) modes. Experimental results indicate that the proposed model with fixed weights approach outperformed the adaptive weights initialization with in relatively minimal number of training epochs. These findings contribute to the development of robust and efficient models for articulatory feature prediction, paving the way for advancements in speech production research and applications. [ABSTRACT FROM AUTHOR]

The Shear wave (S-wave) velocity is an essential parameter in reservoir characterization and evaluation, fluid identification, and prestack inversion. However, the cost of obtaining S-wave velocities directly from dipole acoustic logging is relatively high. At the same time, conventional data-driven S-wave velocity prediction methods exhibit several limitations, such as poor accuracy and generalization of empirical formulas, inadequate exploration of logging curve patterns of traditional fully connected neural networks, and gradient explosion and gradient vanishing problems of recurrent neural networks (RNNs). In this study, we present a reliable and low-cost deep learning (DL) approach for S-wave velocity prediction from real logging data to facilitate the solution of these problems. We designed a new network sensitive to depth sequence logging data using conventional neural networks. The new network is composed of one-dimensional (1D) convolutional, bidirectional long short-term memory (BiLSTM), attention, and fully connected layers. First, the network extracts the local features of the logging curves using a 1D convolutional layer, and then extracts the long-term sequence features of the logging curves using the BiLSTM layer, while adding an attention layer behind the BiLSTM network to further highlight the features that are more significant for S-wave velocity prediction and minimize the influence of other features to improve the accuracy of S-wave velocity prediction. Afterward, the nonlinear mapping relationship between logging data and S-wave velocity is established using several fully connected layers. We applied the new network to real field data and compared its performance with three traditional methods, including a long short-term memory (LSTM) network, a back-propagation neural network (BPNN), and an empirical formula. The performance of the four methods was quantified in terms of their coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE). The new network exhibited better performance and generalization ability, with R² greater than 0.95 (0.9546, 0.9752, and 0.9680, respectively), RMSE less than 57 m/s (56.29, 23.18, and 30.17 m/s, respectively), and MAE less than 35 m/s (34.68, 16.49, and 21.47 m/s, respectively) for the three wells. The test results demonstrate the efficacy of the proposed approach, which has the potential to be widely applied in real areas where S-wave velocity logging data are not available. Furthermore, the findings of this study can help for a better understanding of the superiority of deep learning schemes and attention mechanisms for logging parameter prediction. [ABSTRACT FROM AUTHOR]

Lithium-ion batteries are becoming more popular due to their superior performance like high power density, long lifespan, broad operating range of temperatures, quick charging capabilities, and low self-discharge. The implementation of a Battery Management System (BMS) is crucial in order to guarantee the secure and optimal functioning of electric vehicle batteries. BMS monitors, controls, and maintains the health and performance of rechargeable batteries by monitoring the State of Charge (SoC), State of Health (SoH), and State of Temperature (SoT). SoC quantifies the amount of energy stored in a battery at a certain moment, and it is employed to approximate the remaining distance that may be covered. SoH describe the overall health or state of a rechargeable battery. It reveals how well a battery performs in comparison to its original state. SoT estimation keeps the battery temperature within a safe range, allowing it to reach a higher age and safety, which is critical for battery reliability. In this paper SoC, SoH, SoT estimation models for a lithium-ion battery have been developed using an improved EP-based R110-BLSTM approach. The Emperor Penguin based Residual Network-110 incorporated Bidirectional Long-Short Term Memory (EP-based R110-BLSTM) is ideal for estimating SoC, SoH, and SoT and features great accuracy, a quick estimation speed, and strong generalization capabilities. However, Extreme Learning Machine performance is heavily reliant on proper feature extraction. In order to enhance estimate performance by extracting the best features, Simulated-Annealing-based Golden Eagle optimization is utilized. Electric vehicle drive cycles are used to test the model's resilience to temperature changes. MATLAB 2018b software is used in the execution of this research. The results demonstrate that the proposed model performs better in terms of accuracy and has lower SoC, SoH, and SoT error rates than existing models. A thorough comparison between the recommended model and existing methods is also made, further demonstrating the proposed model's superiority. [ABSTRACT FROM AUTHOR]

Transfer learning and domain adaptation (DA) methods have been utilized in bearing prognostic and health management, but most of the current DA methods do not take into account the feature scale change of degraded features when aligning the feature distribution, and these methods are more suitable for the classification problem, which is more robust to the feature scale change. However, they perform poorly in regression problems. In addition, most of the remaining useful life (RUL) prediction methods require preprocessing such as statistical feature extraction on the signal, which makes the prediction process complicated. To solve the above problems, a DA method based on the representation subspace distance (RSD) is proposed for predicting the bearing RUL under different operating conditions. First, the proposed convolutional neural network (CNN) self-attention (SA) long short term memory network model is utilized to extract the deep features from the original signal, which overcomes the limitations of the CNN in extracting time series. Then, the RSD in the Riemannian geometry of the Grassmann manifold is proposed as a domain transfer loss to learn domain invariant features. The modified method can align the feature distribution of the source domain and the target domain without changing the feature scale. At the same time, the bases mismatch penalization is introduced to avoid destroying the semantic information of the features in the process of domain alignment. Finally, the effectiveness of the proposed method is verified by experiments on four types of transfer tasks, and its superiority is also demonstrated by comparison with other advanced methods. [ABSTRACT FROM AUTHOR]

Opinion and attempts to develop an automated system to determine people's viewpoints towards various units such as events, topics, products, services, organizations, individuals, and issues. Opinion analysis from the natural text can be regarded as a text and sequence classification problem which poses high feature space due to the involvement of dynamic information that needs to be addressed precisely. This paper introduces effective modelling of human opinion analysis from social media data subjected to complex and dynamic content. Firstly, a customized preprocessing operation based on natural language processing mechanisms as an effective data treatment process towards building quality-aware input data. On the other hand, a suitable deep learning technique, bidirectional long short term-memory (Bi-LSTM), is implemented for the opinion classification, followed by a data modelling process where truncating and padding is performed manually to achieve better data generalization in the training phase. The design and development of the model are carried on the MATLAB tool. The performance analysis has shown that the proposed system offers a significant advantage in terms of classification accuracy and less training time due to a reduction in the feature space by the data treatment operation. [ABSTRACT FROM AUTHOR]

The monitoring of human behavior and traffic surveillance in various locations has become increasingly important in recent years. However, identifying abnormal activity in real-world settings is a challenging task due to the many different types of worrisome and abnormal actions, including theft, violence, and accidents. To address this issue, this paper proposes a new framework for deep learning-based anomaly identification in videos using the squirrel search algorithm and bidirectional long short-term memory (BiLSTM). The proposed method combines the squirrel search algorithm, an optimization technique inspired by nature, with BiLSTM for anomaly recognition. The framework uses the knowledge gained from a sequence of frames to categorize the video as either typical or abnormal. The proposed method was exhaustively tested in several benchmark datasets for anomaly detection to confirm its functionality in challenging surveillance circumstances. The results show that the proposed framework outperforms existing methods in terms of area under curve (AUC) values, with a test set AUC score of 93.1%. The paper also discusses the importance of feature selection and the benefits of using BiLSTM over traditional unidirectional long short-term memory (LSTM) models for anomaly detection in videos. Overall, the proposed framework provides a highly precise computerization of the system, making it an effective tool for identifying abnormal human behavior in surveillance footage. [ABSTRACT FROM AUTHOR]

Thyroid classification is required in the medical domain to better assist doctors in deciding diagnostic treatments. Although many researchers conducted experiments to detect abnormal conditions of fetal brains in an earlier stage, there exist several limitations like over-fitting problems and imbalance data problems. To deal with these problems, this paper proposes 'Transfer learning- Bidirectional Long Short Term Memory (TL-BiLSTM) which is an efficient thyroid classification model. This paper focuses on identifying the defects in fetal brains in a primary stage by investigating the thyroid range of the mother during the 19th week of pregnancy. In this research, TL is applied with Bi-LSTM for the improvement of Thyroid classification performance. The Transfer learning method selects the optimal batch size for the Bi-LSTM model to eliminate the overfitting problem. The bi-LSTM model learns the sequence in forward and reverses mode to store the useful features for the long term and discard the irrelevant features. The most significant features in the dataset are selected by applying a modified firefly algorithm (MFA). The modified firefly algorithm has the advantages of easy escape from local optima and a good convergence rate. For evaluation purposes, the thyroid dataset is used as input for investigating the proposed classifier's effectiveness. The evaluation results display that the proposed novel approach successfully identifies and classifies thyroid problems using fetal brain magnetic resonance imaging (MRI) images of various Gestational weeks. [ABSTRACT FROM AUTHOR]

We examine named entity recognition (NER), an essential and commonly used first step in many natural language processing tasks, including chatbots and language translation. We focus on the application of NER to texts that have a lot of noise, such as tweets, which is difficult due to the casual and unstructured language often used in these mediums. In this study, we make use of the largest available labeled data sets for Turkish NER, specifically targeting three informal platforms, namely Twitter, Facebook and Donanimhaber. We choose Turkish as a representative agglutinative language, which has a significantly different structure than other well-known languages such as English, French, and German. We emphasize that the methodologies and insights gained from this study can be extended to other agglutinative languages, like Finnish, Hungarian, Japanese, and Korean. We apply NER to these datasets using 16 different named entity tags through a framework that employs bidirectional long short-term memory (BiLSTM) networks followed by conditional random fields (CRF), known together as the BiLSTM-CRF model. Our experiments show an F1 score of 84% on a combined dataset, which indicates that deep learning models can also be effectively used for business applications in informal settings in agglutinative languages such as Turkish. [ABSTRACT FROM AUTHOR]

Stock index forecasting has always been an interesting subject for investors and related scholars. Accurately stock index forecasting can provide some helpful suggestions for investors and keep financial markets stable. In this study, a new forecasting system, including point prediction and interval prediction, has been proposed to predict the stock index. For obtaining a better predictive effect, multiple influencing variables are also considered in the novel model. More specifically, in the point prediction models, this study applies gradient boosting decision tree (GBDT) to choose some variables related to the stock index by determining their contribution to accurate prediction. Next, an autoencoder (AE) is utilized to reduce the dimensionality of screened factors for the purpose of reducing the effect of noise and improving the efficiency of forecasting. These reconstructed features are all inputted into bidirectional long short-term memory (BiLSTM) to do point prediction. The interval prediction is based on point prediction results and Gaussian process regression (GPR), intended to quantitative uncertainty of the variables. This study chooses the Chinese stock index including the Shanghai Securities Composite Index (SSEC), Shenzhen Composite Index (SZI) and China Securities Index 300 (CSI300) to demonstrate the validity of the innovative hybrid model. Furthermore, this study also selects some other models for comparison. Evaluating the performance of the novel hybrid model, it could be considered as a valid way to do stock index forecasting. [ABSTRACT FROM AUTHOR]

The failure of skin disease detection at an early stage leads to causes of well-known cancer known as melanoma, and it is created due to an assortment of dermatological conditions. Based on morphological attributes, design, surface, and shading, they are isolated into different classifications. To minimize the mortality rate, the early and timely prediction and diagnosis model is essential in medical field; so, to perform automatic detection, a novel hybrid flash butterfly optimized convolutional neural network with bidirectional long short-term memory (HFB-CNN-BiLSTM) approach is to accurately predict and classify the category of skin disease captured from dermoscopic images. The images are gathered from Ham10000 datasets that are highly imbalanced, and during training, it degrades classification performance. Therefore, the images are balanced by using preprocessing pipeline like augmentation by increasing the number of training samples to improve the efficiency of classification performance. Then feature extraction and classification processes are performed using HFB-CNN-BiLSTM to extract the relevant image features and classify them accurately based on their lesion characteristics as normal and abnormal (melanoma, benign keratosis, and melanocytic nevus). Moreover, the proposed framework's viability is examined using MATLAB2018b software, and the performance is validated by comparison with existing approaches for various metrics. As a result, the proposed HFB-CNN-BiLSTM approach is highly superior in terms of all performance metrics compared to other existing approaches. The classification accuracy achieved by the proposed HFB-CNN-BiLSTM model in detecting three kinds of skin diseases is about 96.3%. [ABSTRACT FROM AUTHOR]

Automatic seizure detection plays a key role in assisting clinicians for rapid diagnosis and treatment of epilepsy. In view of the parallelism of temporal convolutional network (TCN) and the capability of bidirectional long short-term memory (BiLSTM) in mining the long-range dependency of multi-channel time-series, we propose an automatic seizure detection method with a novel end-to-end TCN-BiLSTM model in this work. First, raw EEG is filtered with a 0.5–45 Hz band-pass filter, and the filtered data are input into the proposed TCN-BiLSTM network for feature extraction and classification. Post-processing process including moving average filtering, thresholding and collar technique is then employed to further improve the detection performance. The method was evaluated on two EEG database. On the CHB-MIT scalp EEG database, our method achieved a segment-based sensitivity of 94.31%, specificity of 97.13%, and accuracy of 97.09%. Meanwhile, an event-based sensitivity of 96.48% and an average false detection rate (FDR) of 0.38/h were obtained. On the SH-SDU database we collected, the segment-based sensitivity of 94.99%, specificity of 93.25%, and accuracy of 93.27% were achieved. In addition, an event-based sensitivity of 99.35% and a false detection rate of 0.54/h were yielded. The total detection time consumed for 1 h EEG data was 5.65 s. These results demonstrate the superiority and promising potential of the proposed method in real-time monitoring of epileptic seizures. [ABSTRACT FROM AUTHOR]

Entity relation extraction holds a significant role in extracting structured information from unstructured text, serving as a foundational component for various other tasks within natural language processing. The pipeline method in entity relation extraction separates entity subtask from relation subtask, causing an error propagation. Contemporary researchers are more inclined to amalgamate two subtasks, improve and innovate the structures of models to carry out joint entity relation extraction. However, these models often merely capture surface-level text features, overlooking the profound-level semantics and syntax inherent within sentences, consequently forfeiting valuable knowledge. In this condition, we propose a joint entity relation extraction method that integrates context semantic and dependency syntax. The bidirectional long short-term memory network is employed to explore context semantic features of sentences, and tree-structured LSTM is utilized to extract dependency syntactic features, subsequently two types of features are fused with the attention mechanism for joint extraction. Experiment results demonstrate that compared with other models, the Accuracy, Recall and F1-value of our proposed method are increased evidently, proving that semantic and syntactic information contained in sentences are beneficial for entity relation extraction. [ABSTRACT FROM AUTHOR]

Copyright of Bearing is the property of Bearing Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Accurate crop yield prediction is extremely useful to global food production. On the basis of precise forecasts, timely import and export choices should be made. The model of crop yield prediction facilitates the farmers for making better decision regarding the suitable time for crop cultivation. In this study, the prediction of major crops in India is focused by using weather, soli and rainfall data.This study uses pre-processing, feature selection (FS) and prediction model. Initially, the dataset is normalized and the necessary features are selected by three FS models. The FS models are Lasso Based Feature Selection (LFS), Correlation Based Feature Selection (CFS) and Mutual Information Based Feature Selection (MIFS). Then deep learning (DL) based optimization (Attention with Bidirectional Long Short-Term Memory (A-BiLSTM)-MayFlyAlgorithm (MFA) is used for crop prediction. This optimization is used to minimize the loss function; thereby achieving better prediction. In India, the crops like Rice, sugarcane, wheat andmaize are the most cultivatable. Hence, in this work, these crops are considered for prediction. The performance of the BiLSTM- MFA is compared with certain DL models on the basis of error measures. [ABSTRACT FROM AUTHOR]

In today's industrial landscape, the imperative of fault warning for equipment and systems underscores its critical significance in research. The deployment of fault warning systems not only facilitates the early detection and identification of potential equipment failures, minimizing downtime and maintenance costs, but also bolsters equipment reliability and safety. However, the intricacies and non-linearity inherent in industrial data often pose challenges to traditional fault warning methods, resulting in diminished performance, especially with complex datasets. To address this challenge, we introduce a pioneering fault warning approach that integrates an enhanced Coati Optimization Algorithm (ICOA) with a Bidirectional Long Short-Term Memory (Bi-LSTM) network. Our strategy involves a triple approach incorporating chaos mapping, Gaussian walk, and random walk to mitigate the randomness of the initial solution in the conventional Coati Optimization Algorithm (COA). We augment its search capabilities through a dual population strategy, adaptive factors, and a stochastic differential variation strategy. The ICOA is employed for the optimal selection of Bi-LSTM parameters, effectively accomplishing the fault prediction task. Our method harnesses the global search capabilities of the COA and the sophisticated data analysis capabilities of the Bi-LSTM to enhance the accuracy and efficiency of fault warnings. In a practical application to a real-world case of induced draft fan fault warning, our results indicate that our method anticipates faults approximately two hours in advance. Furthermore, in comparison with other advanced methods, namely, the Improved Social Engineering Optimizer Optimized Backpropagation Network (ISEO-BP), the Sparrow Particle Swarm Hybrid Algorithm Optimized Light Gradient Boosting Machine (SSAPSO-LightGBM), and the Improved Butterfly Optimization Algorithm Optimized Bi-LSTM (MSBOA-Bi-LSTM), our proposed approach exhibits distinct advantages and robust prediction effects. [ABSTRACT FROM AUTHOR]

Accurate identification of the backbone curves of reinforced concrete (RC) columns is key to engineering design and strengthening renovation. In view of the problems of high cost, long time, low accuracy, large dispersion of calculation results and discontinuous stiffness changes of the existing backbone curve identification methods, such as experimental method, finite element simulation method and semi-theoretical and semi-empirical method, it is proposed to transform the backbone curve identification problem into a multi-time series prediction problem. By introducing the attention mechanism and combining it with the bidirectional long short-term memory (BiLSTM), the backbone curve identification model (BC-ABiLSTM) is established considering the relationship between the front and back points of the backbone curves. Compared with the models for backbone curves with BiLSTM (BC-BiLSTM), long short-term memory (BC-LSTM), multilayer perceptron (BC-MLP), and the existing identification methods, the performance of the BC-ABiLSTM model is better, and the mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), and R2 of the BC-ABiLSTM model on the testing set are 12.492 kN, 10.595 %, 20.838 kN and 0.9924, respectively, which provides a new method for accurate, efficient and cost-effective prediction of RC column backbone curve points under various cyclic loading levels.

Abstract Background Lysine crotonylation (Kcr) is a crucial protein post-translational modification found in histone and non-histone proteins. It plays a pivotal role in regulating diverse biological processes in both animals and plants, including gene transcription and replication, cell metabolism and differentiation, as well as photosynthesis. Despite the significance of Kcr, detection of Kcr sites through biological experiments is often time-consuming, expensive, and only a fraction of crotonylated peptides can be identified. This reality highlights the need for efficient and rapid prediction of Kcr sites through computational methods. Currently, several machine learning models exist for predicting Kcr sites in humans, yet models tailored for plants are rare. Furthermore, no downloadable Kcr site predictors or datasets have been developed specifically for plants. To address this gap, it is imperative to integrate existing Kcr sites detected in plant experiments and establish a dedicated computational model for plants. Results Most plant Kcr sites are located on non-histones. In this study, we collected non-histone Kcr sites from five plants, including wheat, tabacum, rice, peanut, and papaya. We then conducted a comprehensive analysis of the amino acid distribution surrounding these sites. To develop a predictive model for plant non-histone Kcr sites, we combined a convolutional neural network (CNN), a bidirectional long short-term memory network (BiLSTM), and attention mechanism to build a deep learning model called PlantNh-Kcr. On both five-fold cross-validation and independent tests, PlantNh-Kcr outperformed multiple conventional machine learning models and other deep learning models. Furthermore, we conducted an analysis of species-specific effect on the PlantNh-Kcr model and found that a general model trained using data from multiple species outperforms species-specific models. Conclusion PlantNh-Kcr represents a valuable tool for predicting plant non-histone Kcr sites. We expect that this model will aid in addressing key challenges and tasks in the study of plant crotonylation sites.

Given the wide range of sensor applications, pedestrian activity recognition research using smartphone sensors has gained significant attention. Recognizing activities can yield valuable insights into a person’s actions and the context of the activities. This study proposed a bidirectional long short-term memory based on the feature fusion model with a convolutional neural network (BiLSTM-BFF with CNN) to integrate time and frequency domain features and CNN. The fused feature vector was used as input in the BiLSTM network. The BiLSTM-BFF with CNN model recognized 14 types of pedestrian activity. New pedestrian activity datasets were collected from smartphone sensors used by different types of people (men, women, children, pregnant women, people with limps) and activities (walking, fast walking, elevator up and down, step escalator up and down, walking with step escalator up and down, flat escalator up and down, walking with flat escalator up and down, upstairs and downstairs). The efficiency of the proposed BiLSTM-BFF with the CNN model was validated by conducting experiments using this new dataset. The proposed method demonstrated 95.35% accuracy in recognizing pedestrian activities. The results highlighted the superior accuracy of the proposed method compared to other methods.

Battery energy storage systems, especially lithium-ion batteries, have become more common in power systems owing to their numerous advantages, such as supporting voltage and frequency regulation and contributing to peak shaving and load shifting. However, when the battery reaches its end-of-life, it becomes more unstable, leading to a higher probability of system operation failure and safety accidents. Therefore, to accurately predict the State of Health (SOH) and the Remaining Useful Life (RUL) of a battery system, a prediction method is proposed in this paper based on Empirical Mode Decomposition (EMD), Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Network (CNN), and Attention Mechanism (AM). Firstly, capacity and different health indicators with high correlation extracted from the battery’s charging and discharging characteristics are considered inputs. Then, the EMD method decomposes the battery data into several intrinsic mode functions (IMFs) and a residual. In the second part, with IMFs and a residual as input parameters, the SOH and RUL of different battery datasets are predicted by using the combined model CNN-BiLSTM-AM. To validate the accuracy of the proposed method, different comparative models are considered and carried out on CALCE and NASA battery degradation datasets. The results illustrate that the errors of the proposed method, which are root mean square error and mean absolute error are at least 48% and 19% more accurate than others in all battery datasets, showing the effectiveness and accuracy of the proposed model in predicting the SOH and RUL of the battery.

In recent lip-reading technologies, deep learning methodologies have emerged as the key, transcending the limitations of traditional hybrid Deep Neural Network-Hidden Markov Model (DNN-HMM) frameworks based on Discrete Cosine Transform (DCT) features. LipSyncNet comprises a three-dimensional-Convolutional Neural Network (3D-CNN) that consists of a maximum depth of four layers and is responsible for extracting visual features by integrating EfficientNetB0, which results in excellent feature extraction capabilities. Following this, the network architecture incorporates a backend that utilizes a Bidirectional Long Short-Term Memory (Bi-LSTM)—a component of the recurrent neural network family—combined with Connectionist Temporal Classification (CTC) loss, enhancing its ability to perform classification tasks. The effectiveness of the proposed method is demonstrated through the evaluation of the Graphics Research International Database (GRID) corpus, a challenging word-level lip-reading dataset. Initially, facial features are extracted from the mouth area of an individual’s face. Subsequently, these features are combined with available audio information to identify spoken words precisely. The lip-reading method aims to create a system that achieves accurate speech recognition by observing visual cues, thereby reducing the reliance on audio. The model utilizes information from various levels in a unified structure, enabling it to differentiate between words that sound alike and to improve its ability to handle changes in physical appearance.

Detecting anomalous executions in business process data is crucial for safeguarding the efficiency and success of an organization. Unsupervised approaches are commonly used for business process anomaly detection because of the scarcity of labeled anomaly data. However, these approaches often encounter a notable decline in performance because they lack prior knowledge about the anomalies. Additionally, most of them do not perform root cause analysis on the detected anomalies. This study proposes a variational autoencoder-based approach to overcome the performance limitations of existing unsupervised methods and determine the root causes of the detected anomalies. The learning of the variational autoencoder from unlabeled business process data is enhanced in the proposed approach by leveraging different architectural components, namely, the entity embedding technique, the bidirectional long short-term memory network, and the self-attention mechanism. Combining these architectural components in the variational autoencoder architecture leads to learning high-level representations from the business process data and thus improving the reconstruction capability of the variational autoencoder. Furthermore, this study suggests feeding the reconstruction error provided by the variational autoencoder into the logistic regression classifier to improve the accuracy of anomaly detection. The performance of the proposed model was evaluated on real-life and synthetic datasets. The experimental findings indicate that the proposed model outperforms six existing anomaly detection models in terms of precision, recall, and F1-score metrics.

Long-term failures in flexible converter valve equipment pose significant risks, potentially compromising operational efficiency or leading to complete malfunction. Accurately identifying equipment faults is essential to improve overall reliability and minimize downtime. This study introduces an innovative fault diagnosis method utilizing an attention mechanism. The method integrates a lightweight model incorporating one-dimension depthwise convolutional layers for spatial feature extraction and bidirectional long short-term memory for capturing temporal dynamics. A pioneering time-channel joint attention module enhances the extraction of fault-related data from time series and channel maps. Experimental results underscore the method’s efficacy in fault diagnosis under varying Gaussian noise conditions. Notably, the approach demonstrates remarkable consistency in accuracy across various experimental setups, underscoring its robust performance and potential applicability in real-world scenarios where reliability is critical. In addition, the proposed method has a moderate number of parameters and training time, indicating that the model can be embedded in front-end equipment.

Rolling bearings are essential in the industrial field as a critical component of mechanical systems. Therefore, accurately predicting the remaining useful life of rolling bearings is vital to the safety and reliability of mechanical operation. However, traditional life prediction methods often have problems such as insufficient feature extraction and poor model generalization capabilities, which lead to more significant errors. To solve the above problems, this paper proposes a novel remaining useful life (RUL) prediction method of rolling bearings based on integrated multi-head attention (MHA), improved temporal convolutional network (TCN), and bidirectional long short-term memory (BiLSTM). This method utilizes an improved TCN-BiLSTM network to capture dependencies in sequences and extract global features from signals. In the meantime, MHA is introduced to fully capture the degradation information of the bearing and ultimately predict the life of the bearing. Finally, the bearing life prediction process is fully demonstrated through novel three-dimensional feature visualization. To verify the effectiveness of this method, this paper conducted RUL prediction experiments using the IEEE PHM 2012 dataset and the XJTU-SY dataset, respectively. Many experiments are organized to test the performance, and the experimental results show that this method has higher prediction accuracy and robustness than other methods.