Bidirectional Long Short-term Memory Layers Research Articles

ENHANCING 〖PM〗_(2.5) PREDICTION IN KEMAYORAN DISTRICT, DKI JAKARTA USING DEEP BILSTM METHOD

Worldwide air pollution is a concern, and this is especially true in Indonesia, where most people breathe air that is more contaminated than recommended by the WHO. The concentration of presents notable health hazards. The respiratory system is the primary route of absorption for , allowing it to enter the lung alveoli and enter the bloodstream. Given the significant health risks associated with exposure, accurate forecasting methods are crucial to anticipate and mitigate its effects. Traditional forecasting methods like ARIMA have limitations in handling non-linear and complex patterns. Therefore, an accurate machine learning method is needed to improve forecasting performance. This research employs Deep Bidirectional Long-Short Term Memory (BiLSTM), a deep learning model particularly suited for time series forecasting due to its ability to capture both past and future dependencies in sequential data. To achieve accurate and precise forecasts for predicting concentration levels in Kemayoran District in November , 2023 (24 hours), this research utilized hourly concentration data from May until October , 2023, using Deep BiLSTM. The outcomes demonstrated the efficiency of the model, attaining a Mean Absolute Percentage Error (MAPE) of 17.1540% (training) and 14.2862% (testing) with an 80:20 data split. The optimal parameters, which comprised 24 timesteps, Adam optimizers with a learning rate of 0.001, 16 batch sizes, 1000 epochs, and ReLU activation functions across multiple BiLSTM layers, showcased the model’s effectiveness in forecasting the concentration in Kemayoran District, DKI Jakarta, on November , 2023.

Advancements in Nepali Speech Recognition: A Comparative Study of BiLSTM, Transformer, and Hybrid Models

In today's world, leveraging Automatic Speech Recognition (ASR) technology to process and understand spoken language is highly desirable. Our proposed Nepali Speech Recognition employs an advanced generation to recognize and interpret spoken Nepali language. It approaches Nepali speech, allowing it to reply to user queries effectively. To attain this, we rent a mixture of superior neural network fashions. We extract Mel-frequency cepstral coefficients (MFCCs) from the preprocessed audio information; these MFCCs capture crucial spectral characteristics of Nepali speech and serve as essential input features for our neural network model. To design a top-rated version for textual content-based query processing, we make use of convolutional neural networks (CNN), residual networks (ResNet), and bidirectional long short-term memory (BiLSTM) layers. The CNN layers excel at extracting neighborhood patterns and spatial features from the MFCC input; the ResNet layers capture deeper representations to enhance performance. The BiLSTM layers are also employed to model temporal dependencies in the textual content-based query processing, we make use of convolutional neural networks (CNN), residual networks (ResNet), and bidirectional long short-term memory (BiLSTM) layers. The CNN layers excel at extracting neighborhood patterns and spatial features from the MFCC input; the ResNet layers capture deeper representations to enhance performance. The BiLSTM layers are also employed to model temporal dependencies in the textual content records. We hired the Connectionist Temporal classification (CTC) loss feature to enable sequence-to-series mapping, aligning the input speech with corresponding text outputs. This approach permits our gadget to successfully process textual content queries and provide correct responses, enhancing the user's usefulness. The model, after being trained with 1.55 million parameters in about 1 lakh 57 thousand audio datasets for 47 epochs, achieved a CTC of 17.98% (82.02%-character accuracy rate) with this model.

Air quality prediction using stacked bi- long short-term memory and convolutional neural network in India

Air quality is a critical aspect of environmental health, and its assessment and prediction serve as pivotal components in mitigating the adverse effects of air pollution. This study focuses on advancing air quality prediction in India through the application of cutting-edge deep learning techniques, specifically the Stacked Bidirectional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Network (CNN) architecture. Through meticulous preprocessing - encompassing missing value handling, normalization, and temporal sequencing - the dataset is prepared for the Stacked Bi-LSTM and CNN hybrid model. The model architecture leverages the temporal sequence-capturing capabilities of Stacked Bi-LSTM layers, enhancing it with the spatial feature extraction prowess of CNN layers. This integrated approach aims to address the intricate and nonlinear dependencies present in air quality time series data. During the training phase, the Adam optimizer is used to fine-tune the model’s hyperparameters, with Mean Squared Error (MSE) serving as the loss function. Important assessment metrics, including as Mean Absolute Percentage Error (MAPE), Root Mean Squared Error (RMSE), and MSE, are used to evaluate the performance of the model. Furthermore, this study conducts a detailed temporal analysis, unraveling diurnal, seasonal, and long-term trends in air quality fluctuations. The study aims to offer valuable insights into the temporal and spatial patterns of air quality in India, thereby aiding environmental policymakers, urban planners, and researchers in formulating effective strategies for air quality management. The application of Stacked Bi-LSTM and CNN architectures in this research holds promise for enhancing real-time forecasting accuracy and facilitating informed decision-making towards sustainable environmental practices.

Advance drought prediction through rainfall forecasting with hybrid deep learning model

Drought is a natural disaster that can affect a larger area over time. Damage caused by the drought can only be reduced through its accurate prediction. In this context, we proposed a hybrid stacked model for rainfall prediction, which is crucial for effective drought forecasting and management. In the first layer of stacked models, Bi-directional LSTM is used to extract the features, and then in the second layer, the LSTM model will make the predictions. The model captures complex temporal dependencies by processing multivariate time series data in both forward and backward directions using bi-directional LSTM layers. Trained with the Mean Squared Error loss and Adam optimizer, the model demonstrates improved forecasting accuracy, offering significant potential for proactive drought management.

Bearing Fault Diagnosis Method Based on Osprey–Cauchy–Sparrow Search Algorithm-Variational Mode Decomposition and Convolutional Neural Network-Bidirectional Long Short-Term Memory

To solve the problem of the low diagnosis rate of early weak faults of rolling bearings, a novel bearing fault diagnosis method based on Variational Mode Decomposition (VMD) and convolutional neural network (CNN)−Bidirectional Long Short-Term Memory (BiLSTM) was proposed. Based on the basic Sparrow Search Algorithm, the tent chaotic mapping, the Osprey Optimization Algorithm, and the Cauchy mutation were used to enhance the global search ability of the algorithm. To improve the accuracy of fault diagnosis, the BiLSTM layer is introduced into CNN to preserve the global and local features to the maximum extent. The experimental results show that VMD avoids the end effect problem in Empirical Mode Decomposition (EMD). The accuracy rate of the diagnosis model based on CNN-BILSTM reached 97.6667%, which was higher than that of the common diagnosis model.

Empowering Urdu sentiment analysis: an attention-based stacked CNN-Bi-LSTM DNN with multilingual BERT

Sentiment analysis (SA) as a research field has gained popularity among the researcher throughout the globe over the past 10 years. Deep neural networks (DNN) and word vector models are employed nowadays and perform well in sentiment analysis. Among the different deep neural networks utilized for SA globally, Bi-directional long short-term memory (Bi-LSTM), BERT, and CNN models have received much attention. Even though these models can process a wide range of text types, Because DNNs treat different features the same, using these models in the feature learning phase of a DNN model leads to the creation of a feature space with very high dimensionality. We suggest an attention-based, stacked, two-layer CNN-Bi-LSTM DNN to overcome these glitches. After local feature extraction, by applying stacked two-layer Bi-LSTM, our proposed model extracted coming and outgoing sequences by seeing sequential data streams in backward and forward directions. The output of the stacked two-layer Bi-LSTM is supplied to the attention layer to assign various words with varying values. A second Bi-LSTM layer is constructed atop the initial layer in the suggested network to increase performance. Various experiments have been conducted to evaluate the effectiveness of our proposed model on two Urdu sentiment analysis datasets named as UCSA-21 and UCSA, and an accuracies of 83.12% and 78.91% achieved, respectively.

Enhancing Lung Cancer Screening with Bidirectional LSTM and GRU Models

Abstract. This study aims to enhance lung cancer patient screening by developing and evaluating bidirectional Long Short-Term Memory (LSTM) and bidirectional Gated Recurrent Unit (GRU) models using the Lung Cancer dataset from Kaggle. The dataset includes 16 features related to patient symptoms and lung cancer status, providing a broad spectrum of symptoms to improve model accuracy. The research advances Artificial Intelligence (AI)-driven healthcare by integrating these sophisticated machine learning techniques into diagnostic processes. The methodology involves four main steps: preprocessing the dataset for model compatibility, defining the model architecture with bidirectional LSTM and GRU layers and evaluating its performance. The results show an overall accuracy of 52.17%, with accuracy, recall, and F1 scores for both cancerous and non-cancerous categories around 50%. Despite the hybrid model's average performance, it establishes a basis for future enhancements. Optimizing model parameters and exploring additional other techniques to improve prediction accuracy and clinical applicability will be done in the future.

Novel cost-effective method for forecasting COVID-19 and hospital occupancy using deep learning

The emergence of the COVID-19 pandemic in 2019 and its rapid global spread put healthcare systems around the world to the test. This crisis created an unprecedented level of stress in hospitals, exacerbating the already complex task of healthcare management. As a result, it led to a tragic increase in mortality rates and highlighted the urgent need for advanced predictive tools to support decision-making. To address these critical challenges, this research aims to develop and implement a predictive system capable of predicting pandemic evolution with accuracy (in terms of Mean Absolute error (MAE), Root Mean Square Error (RMSE), R2, and Mean Absolute Percentage Error (MAPE)) and low computational and economic cost. It uses a set of interconnected Long Short Term-memory (LSTM) with double bidirectional LSTM (BiLSTM) layers together with a novel preprocessing based on future time windows. This model accurately predicts COVID-19 cases and hospital occupancy over long periods of time using only 40% of the set to train. This results in a long-term prediction where each day we can query the cases for the next three days with very little data. The data utilized in this analysis were obtained from the “Hospital Insular” in Gran Canaria, Spain. These data describe the spread of the coronavirus disease (COVID-19) from its initial emergence in 2020 until March 29, 2022. The results show an improvement in MAE (< 161), RMSE (< 405), and MAPE (> 0.20) compared to other studies with similar conditions. This would be a powerful tool for the healthcare system, providing valuable information to decision-makers, allowing them to anticipate and strategize for possible scenarios, ultimately improving public health outcomes and optimizing the allocation of healthcare and economic resources.

Prediction of QT interval in ECGs through baseline wander removal and deep learning technique

Abstract Background In electrocardiograms (ECGs), the QT interval, calculated from the onset of the Q-wave to the end of the T-wave, reflects the repolarization status of the heart. Currently, the QT interval automatically measured in 12-lead electrocardiograms is inaccurately derived as the median value, rather than utilizing the more precise mean value of QT intervals from all leads. This approach fails to provide information on heterogeneous QT prolongation crucial for predicting arrhythmias, as it does not accurately represent the values of QT intervals in each lead. Purpose Traditional analysis of ECGs based on QT intervals necessitates direct interpretation by cardiologists, leading to potential errors in diagnosis. In response to this challenge, we present a model that integrates low-pass filtering and continuous wavelet transformation to eliminate baseline wander noise occurring during ECG measurements. Moreover, to account for temporal features, we incorporate Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) to estimate QT intervals. Methods The dataset used in this study was the QT Database, comprising ECG data for 105 patients recorded at a sampling rate of 250Hz over a duration of 15 minutes. Each ECG was associated with annotations for two leads and information regarding the P, QRS, and T waves. We structured the dataset into a ratio of approximately 8:1:1, resulting in 51 patients for training (15,490 samples), 6 patients for validation (1,564 samples), and 7 patients for testing (1,682 samples). Results For the evaluation of deep learning, we selected Accuracy and F1-Score as metrics. We experimented with two configurations: the proposed model with two Bi-LSTM layers and an alternative with a single Bi-LSTM layer of 100 units. We also tested a similar recurrent neural network, GRU, with the same unit configuration. The results indicated that the method with two Bi-LSTM layers outperformed others in both Accuracy and F1-Score. Conclusion By incorporating CNN and Bi-LSTM, we conducted QT interval estimation, leveraging temporal features. This methodology facilitated the elimination of diverse noise types during ECG measurements, resulting in enhanced accuracy for QT interval estimation through the consideration of temporal ECG characteristics. In our future research endeavors, we plan to deploy this model for predicting arrhythmias, specifically targeting conditions like long QT syndrome and drug-induced QT prolongation.Methods and result

Enhanced UrduAspectNet: Leveraging Biaffine Attention for superior Aspect-Based Sentiment Analysis

Urdu, with its rich linguistic complexity, poses significant challenges for computational sentiment analysis. This study presents an enhanced version of UrduAspectNet, specifically designed for Aspect-Based Sentiment Analysis (ABSA) in Urdu. We introduce key innovations including the incorporation of Biaffine Attention into the model architecture, which synergizes XLM-R embeddings, a bidirectional LSTM (BiLSTM), and dual Graph Convolutional Networks (GCNs). Additionally, we utilize dependency parsing to create the adjacency matrix for the GCNs, capturing syntactic dependencies to enhance relational representation. The improved model, termed Enhanced UrduAspectNet, integrates POS and lemma embeddings, processed through BiLSTM and GCN layers, with Biaffine Attention enhancing the extraction of intricate aspect and sentiment relationships. We also introduce the use of BIO tags for aspect term identification, improving the granularity of aspect extraction. Experimental results demonstrate significant improvements in both aspect extraction and sentiment classification accuracy. This research advances Urdu sentiment analysis and sets a precedent for leveraging sophisticated NLP techniques in underrepresented languages.

AN OPTIMIZED BIDIRECTIONAL CONVOLUTIONAL RECURRENT NEURAL NETWORK ARCHITECTURE WITH GROUP-WISE ENHANCEMENT MECHANISM OF SENTIMENTS FOR THE PERSPECTIVE OF CUSTOMER REVIEW SUMMARIZATION

Customer reviews play pivotal roles in consumers’ purchase decisions, but the sheer volume of text data can be overwhelming. In existing system, while ensemble methods can enhance performance, the associated computational complexity and resource intensiveness should be carefully considered, and appropriate measures should be taken to address these challenges in the context of Customer Review Summarization. This study introduces a Particle Swarm Optimization (PSO) based optimized architecture for a Bidirectional Convolutional Recurrent Neural Network (BiCRNN) with a group-wise enhancement mechanism tailored for Customer Review Summarization and named as OBiCRNN. This model is designed for the perspective of customer review summarization, aiming to effectively capture sentiments and generate concise summaries. The integration of PSO optimizes the network parameters, enhancing the learning process. Feature extraction is done by Modified Principal Component Analysis (MPCA) which uses correlated feature sets and extracts most informative features for given datasets. BiCRNN utilizes bidirectional LSTM and GRU layers for comprehensive context understanding, while the group-wise enhancement mechanism categorizes sentiment-related features, amplifying essential sentiments and attenuating less relevant ones. With this novel approach, the architecture leverages both PSO and BiCRNN for an advanced framework in customer review summarization where outcomes demonstrate the effectiveness of the deep learning (DL) model in producing coherent and informative summaries, enhancing the accessibility of customer feedback for both consumers and businesses. The study contributes to the field of natural language processing (NLP) and customer sentiment analysis, offering a scalable solution for managing the wealth of information present in online customer reviews.

Calibration of Typhoon Track Forecasts Based on Deep Learning Methods

An accurate forecast of typhoon tracks is crucial for disaster warning and mitigation. However, existing numerical weather prediction models, such as the Weather Research and Forecasting (WRF) model, still exhibit significant errors in track forecasts. This study aims to improve forecast accuracy by correcting WRF-forecasted tracks using deep learning models, including Bidirectional Long Short-Term Memory (BiLSTM) + Convolutional Long Short-Term Memory (ConvLSTM) + Wide and Deep Learning (WDL), BiLSTM + Convolutional Gated Recurrent Unit (ConvGRU) + WDL, and BiLSTM + ConvLSTM + Extreme Deep Factorization Machine (xDeepFM), with a comparison to the Kalman Filter. The results demonstrate that the BiLSTM + ConvLSTM + WDL model reduces the 72 h track prediction error (TPE) from 255.18 km to 159.23 km, representing a 37.6% improvement over the original WRF model, and exhibits significant advantages across all evaluation metrics, particularly in key indicators such as Bias2, Mean Squared Error (MSE), and Sequence. The decomposition of MSE further validates the importance of the BiLSTM, ConvLSTM, WDL, and Temporal Normalization (TN) layers in enhancing the model’s spatio-temporal feature-capturing ability.

A fault detection of aero-engine rolling bearings based on CNN-BiLSTM network integrated cross-attention

Abstract Aero-engine rolling bearings are essential for engine health, in which disruptive failures can be prevented and reduce great losses in air flight. To improve the efficiency of fault detection, an improved network, named CNN- BiLSTM -Cross-Attention (CBLCA) was proposed. The Bidirectional Long Short-Term Memory (BiLSTM) layer captures the temporal features as the input data. The cross-attention mechanism is integrated with the Convolutional Neural Networks (CNN) layer and the BiLSTM layer respectively. More important feature information can be identified with the CBLCA model. The proposed model was also validated with the open-sourced aero-engine rolling bearings data set. To improve the identification accuracy, a novel method that combines Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) is used for the data preprocessing. Each original signal sample is transformed into a feature set containing richer information, and the number of features significantly increased in the entire dataset. Compared with some existing LSTM models, such as LSTM, BiLSTM, CNN-BiLSTM, and CNN-LSTM, the classification accuracy was increased by 55%, 54%, 5%, and 7%, respectively. The processing method for vibration signals and the CBLCA model can improve the accuracy and reliability of fault diagnosis for aero-engine rolling bearings.

DeepTimeNet: A novel architecture for precise surface temperature estimation of lithium-ion batteries across diverse ambient conditions

With the growing demand for battery-powered devices and electric vehicles, the need for improved battery performance and safety is paramount. A key determinant of battery health is the accurate monitoring of surface temperature (ST). Conventional ST estimation often depends on direct sensor measurements, which may not be cost-effective and can impact system reliability. This paper presents DeepTimeNet, a novel approach leveraging deep learning (DL) architectures for sensorless ST prediction in lithium-ion batteries. DeepTimeNet combines Convolutional Neural Networks (CNN), ResNet blocks, Inception modules, Bidirectional LSTM, and GRU layers to precisely model the time-dependent behaviour of batteries. A comprehensive evaluation against traditional models, across temperatures ranging from -20 °C to 25 °C and under various driving profiles, including US06 and Urban Dynamometer Driving Schedule (UDDS), is conducted. DeepTimeNet’s performance is quantified by metrics such as mean absolute error (MAE), surpassing that of models like Gated Recurrent Unit-Recurrent Neural Network (GRU-RNN), Convolutional Neural Network-Long Short Term Memory Network (CNN-LSTM), and Long Short Term Memory Network (LSTM). The results demonstrate DeepTimeNet’s superior performance, with an RMSE of 0.0971, MSE of 0.0099, MAE of 0.0912, and MAXE of 0.3963, validating it as an advanced tool for enhancing the efficacy of battery management systems and underscoring its potential as a benchmark for future innovations.

A novel lithium-ion battery state-of-health estimation method for fast-charging scenarios based on an improved multi-feature extraction and bagging temporal attention network

Accurately estimating the state of health (SOH) of fast-charging lithium-ion batteries is crucial for safely and reliably operating battery systems. However, handling data scarcity and rapid charging scenarios under diverse operational conditions is challenging. In this paper, a novel approach for estimating the SOH of lithium-ion batteries (LIBs) is introduced based on an improved multisegment feature extraction and bagging temporal attention network. First, four health features, including the time differences observed at equal voltages, the cumulative integral of voltage changes, the total circuit charge variation and the levels of slope peaks, are identified. Second, a residual bidirectional long short-term memory (Bi-LSTM) attention mechanism is designed to focus on the temporal dimension of battery data by incorporating a multilayered complex neural network design comprising convolutional layers, pooling layers, Bi-LSTM layers and fully connected layers. This design effectively captures the features and relationships contained in battery data. The predictions derived from randomly initialized parameters and multiple submodels are stacked to improve the generalizability of the model. Finally, comprehensive evaluations are conducted through comparative experiments, ablation studies and noise experiments, with six evaluation metrics considered across three datasets. The MAE, RMSE, MAEP and MAXE of the proposed model reach 0.366, 0.605, 0.519 and 1.56, respectively. The results indicate that the proposed method enhances the robustness, resilience and generalizability of the estimates produced under different conditions and noise scenarios.

Optimizing hybrid deep learning models for drug‐target interaction prediction: A comparative analysis of evolutionary algorithms

AbstractIn the realm of Drug‐Target Interaction (DTI) prediction, this research investigates and contrasts the efficacy of diverse evolutionary algorithms in fine‐tuning a sophisticated hybrid deep learning model. Recognizing the critical role of DTI in drug discovery and repositioning, we tackle the challenges of binary classification by reframing the problem as a regression task. Our focus lies on the Convolution Self‐Attention Network with Attention‐based bidirectional Long Short‐Term Memory Network (CSAN‐BiLSTM‐Att), a hybrid model combining convolutional neural network (CNN) blocks, self‐attention mechanisms, and bidirectional LSTM layers. To optimize this complex model, we employ Differential Evolution (DE), Particle Swarm Optimization (PSO), Memetic Particle Swarm Optimization Algorithm (MPSOA), Fire Hawk Optimization (FHO), and Artificial Hummingbird Algorithm (AHA). Through thorough comparative analysis, we evaluate the performance of these evolutionary algorithms in enhancing the CSAN‐BiLSTM‐Att model's effectiveness. By examining the strengths and weaknesses of each algorithm, our study aims to provide valuable insights into DTI prediction, identifying the most effective evolutionary algorithm for hyperparameter tuning in advanced deep learning models. Notably, Fire‐hawk optimization (FHO) emerges as particularly promising, achieving the highest Concordance Index (C‐index) as 0.974 for KIBA datasets and 0.894 for DAVIS datasets and demonstrating exceptional accuracy in ranking continuous predictions across both the datasets.

Improving Bus Passenger Flow Prediction Using Bi-LSTM Fusion Model and SMO Algorithm

Bus passenger flow prediction integrates data analytics and modelling techniques to forecast the number of passengers using bus services, incorporating historical usage patterns, demographics, weather, and events for optimal scheduling and resource allocation. The Bi-LSTM fusion model enhances accuracy by processing past and future features simultaneously, leveraging bidirectional LSTM layers and an attention mechanism to capture temporal dependencies. This approach not only refines insights crucial for urban mobility challenges like traffic management and demand forecasting but also improves route planning and service efficiency. The SMO algorithm initializes with a diverse spider monkey population exploring solution spaces. Through local and global leader phases, it iteratively updates positions based on fitness and probabilistic selections, maintaining a balance between exploration and exploitation. Perturbation-based updates in the local leader phase ensure adaptability, preventing premature convergence, while the global leader phase guides towards better solutions, enhancing efficiency in complex optimization tasks and promoting dynamic adaptation. In Dataset 1, the proposed model achieved a training time of 137 seconds, slightly longer than HA (115s), SARIMA (112s), GRU (123s), and ST-ResNet (113s). It demonstrated superior accuracy at 89%, surpassing HA (66%), SARIMA (68%), GRU (63%), DeepST (78%), and ST-ResNet (84%). In Dataset 2, the model exhibited the lowest RMSE, MAE, and MAPE%, indicating superior predictive accuracy over SVR, CNN, GCN, LSTM, and CONV LSTM models. These findings validate the proposed model's effectiveness in enhancing predictive capabilities for transit forecasting, underscoring its potential to optimize urban mobility and transportation management strategies significantly.

Optimizing hybrid neural networks for precise COVID-19 mRNA vaccine degradation prediction

Conventional hybrid models often miss an essential factor that can lead to less effective performance: intrinsic sequence dependence when combining various neural network (NN) architectures. This study addresses this issue by highlighting the importance of sequence hybridization in NN architecture integration, aiming to improve model effectiveness. It combines NN layers—dense, long short-term memory (LSTM), and gated recurrent unit (GRU)—using the Keras Sequential API for defining the architecture. To provide better context, bidirectional LSTM (BiLSTM) and bidirectional GRU (BiGRU) replace their unidirectional counterparts, enhancing the models through bidirectional structures. Out of 25 NN models tested, 18 four-layer hybrid NN models consist of one-quarter dense layer and the rest BiLSTM and BiGRU layers. These hybrid NN models undergo supervised learning regression analysis, with mean column-wise root mean square error (MCRMSE) as the performance metric. The results show that each hybrid NN model produces unique outcomes based on its specific hybrid sequence. The Hybrid_LGSS model performs better than existing three-layer BiLSTM networks in predictive accuracy and shows lower overfitting (MCRMSEs of 0.0749 and 0.0767 for training and validation, respectively). This indicates that the optimal hybridization sequence is crucial for achieving a balance between performance and simplicity. In summary, this research could help vaccinologists develop better mRNA vaccines and provide data analysts with new insights for improvement.

Deep Learning Based Lipreading for Video Captioning

Visual speech recognition, often referred to as lipreading, has garnered significant attention in recent years due to its potential applications in various fields such as human-computer interaction, accessibility technology, and biometric security systems. This paper explores the challenges and advancements in the field of lipreading, which involves deciphering speech from visual cues, primarily movements of the lips, tongue, and teeth. Despite being an essential aspect of human communication, lipreading presents inherent difficulties, especially in noisy environments or when contextual information is limited. The McGurk effect, where conflicting audio and visual cues lead to perceptual illusions, highlights the complexity of lipreading. Human lipreading performance varies widely, with hearing-impaired individuals achieving relatively low accuracy rates. Automating lipreading using machine learning techniques has emerged as a promising solution, with potential applications ranging from silent dictation in public spaces to biometric authentication systems. Visual speech recognition methods can be broadly categorized into those that focus on mimicking words and those that model visemes, visually distinguishable phonemes. While word-based approaches are suitable for isolated word recognition, viseme-based techniques are better suited for continuous speech recognition tasks. This study proposes a novel deep learning architecture for lipreading, leveraging Conv3D layers for spatiotemporal feature extraction and bidirectional LSTM layers for sequence modelling. The proposed model demonstrates significant improvements in lipreading accuracy, outperforming traditional methods on benchmark datasets. The practical implications of automated lipreading extend beyond accessibility technology to include biometric identity verification, security surveillance, and enhanced communication aids for individuals with hearing impairments. This paper provides insights into the advancements, challenges, and future directions of visual speech recognition research, paving the way for innovative applications in diverse domains.

Predicting Transcription Factor Binding Sites with Deep Learning.

Prediction of binding sites for transcription factors is important to understand how the latter regulate gene expression and how this regulation can be modulated for therapeutic purposes. A consistent number of references address this issue with different approaches, Machine Learning being one of the most successful. Nevertheless, we note that many such approaches fail to propose a robust and meaningful method to embed the genetic data under analysis. We try to overcome this problem by proposing a bidirectional transformer-based encoder, empowered by bidirectional long-short term memory layers and with a capsule layer responsible for the final prediction. To evaluate the efficiency of the proposed approach, we use benchmark ChIP-seq datasets of five cell lines available in the ENCODE repository (A549, GM12878, Hep-G2, H1-hESC, and Hela). The results show that the proposed method can predict TFBS within the five different cell lines very well; moreover, cross-cell predictions provide satisfactory results as well. Experiments conducted across cell lines are reinforced by the analysis of five additional lines used only to test the model trained using the others. The results confirm that prediction across cell lines remains very high, allowing an extensive cross-transcription factor analysis to be performed from which several indications of interest for molecular biology may be drawn.