Gated Recurrent Unit Research Articles

Breast cancer classification based on hybrid CNN with LSTM model

Breast cancer (BC) is a global problem, largely due to a shortage of knowledge and early detection. The speed-up process of detection and classification is crucial for effective cancer treatment. Medical image analysis methods and computer-aided diagnosis can enhance this process, providing training and assistance to less experienced clinicians. Deep Learning (DL) models play a great role in accurately detecting and classifying cancer in the huge dataset, especially when dealing with large medical images. This paper presents a novel hybrid model of DL models combined a Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) for binary breast cancer classification on two datasets available at the Kaggle repository. CNNs extract mammographic features, including spatial hierarchies and malignancy patterns, whereas LSTM networks characterize sequential dependencies and temporal interactions. Our method combines these structures to improve classification accuracy and resilience. We compared the proposed model with other DL models, such as CNN, LSTM, Gated Recurrent Units (GRUs), VGG-16, and RESNET-50. The CNN-LSTM model achieved superior performance with accuracies of 99.17% and 99.90% on the respective datasets. This paper uses prediction evaluation metrics such as accuracy, sensitivity, specificity, F-score, and the AUC curve. The results showed that our model CNN-LSTM can enhance the performance of breast cancer classifiers compared with others with 99.90% accuracy on the second dataset.

An Enhanced LSTM Approach for Detecting IoT-Based DDoS Attacks Using Honeypot Data

One of the widening perils in network security is the Distributed Denial of Service (DDoS) attacks on the Internet of Things (IoT) ecosystem. This paper presents an enhanced Intrusion Detection System (IDS) through the proposal of an enhanced version of the long short-term memory (LSTM) model to detect DDoS attacks using honeypot-generated data. The proposed model aggregates the Conv1D, Bidirectional Long Short-Term Memory (Bi-LSTM), Bidirectional Gated Recurrent Unit (Bi-GRU), and dropout layers to extract temporal and spatial features from IoT traffic effectively. We tested the efficacy of the proposed system on a real-world IoT-DH dataset, which showed a remarkable accuracy of 99.41%, with an AUC score of 0.9999. A comparative analysis with other baseline models, such as LSTM, Bidirectional LSTM (Bi-LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Feedforward Neural Network (FNN), and Temporal Convolutional Network (TCN), proved that enhanced LSTM outperformed the other models. This indicates the robustness of the proposed model in correctly detecting DDoS attacks with high generalization capability for unseen traffic data. The contribution of this paper will be an addition to the deep learning techniques applied for the solution of intrusion detection systems (IDS), which will also allow the building and implementation of more efficient security mechanisms in IoT environments.

Federated Learning for IoMT-Enhanced Human Activity Recognition with Hybrid LSTM-GRU Networks

The proliferation of wearable sensors and mobile devices has fueled advancements in human activity recognition (HAR), with growing importance placed on both accuracy and privacy preservation. In this paper, the author proposes a federated learning framework for HAR, leveraging a hybrid Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model to enhance feature extraction and classification in decentralized environments. Utilizing three public datasets—UCI-HAR, HARTH, and HAR7+—which contain diverse sensor data collected from free-living activities, the proposed system is designed to address the inherent privacy risks associated with centralized data processing by deploying Federated Averaging for local model training. To optimize recognition accuracy, the author introduces a dual-feature extraction mechanism, combining convolutional blocks for capturing local patterns and a hybrid LSTM-GRU structure to detect complex temporal dependencies. Furthermore, the author integrates an attention mechanism to focus on significant global relationships within the data. The proposed system is evaluated on the three public datasets—UCI-HAR, HARTH, and HAR7+—achieving superior performance compared to recent works in terms of F1-score and recognition accuracy. The results demonstrate that the proposed approach not only provides high classification accuracy but also ensures privacy preservation, making it a scalable and reliable solution for real-world HAR applications in decentralized and privacy-conscious environments. This work showcases the potential of federated learning in transforming human activity recognition, combining advanced feature extraction methodologies and privacy-respecting frameworks to deliver robust, real-time activity classification.

Enhancing breast cancer prediction through stacking ensemble and deep learning integration

Breast cancer is one of the most common types of cancer in women and is recognized as a serious global public health issue. The increasing incidence of breast cancer emphasizes the importance of early detection, which enhances the effectiveness of treatment processes. In addressing this challenge, the importance of machine learning and deep learning technologies is increasingly recognized. The aim of this study is to evaluate the integration of ensemble models and deep learning models using stacking ensemble techniques on the Breast Cancer Wisconsin (Diagnostic) dataset and to enhance breast cancer diagnosis through this methodology. To achieve this, the efficacy of ensemble methods such as Random Forest, XGBoost, LightGBM, ExtraTrees, HistGradientBoosting, AdaBoost, GradientBoosting, and CatBoost in modeling breast cancer diagnosis was comprehensively evaluated. In addition to ensemble methods, deep learning models including convolutional neural network (CNN), recurrent neural network (RNN), gated recurrent unit (GRU), bidirectional long short-term memory (BILSTM), long short-term memory (LSTM) were analyzed as meta predictors. Among these models, CNN stood out for its high accuracy and rapid training time, making it an ideal choice for real-time diagnostic applications. Finally, the study demonstrated how breast cancer prediction was enhanced by integrating a set of base predictors, such as LightGBM, ExtraTrees, and CatBoost, with a deep learning-based meta-predictor, such as CNN, using stacking ensemble methodology. This stacking integration model offers significant potential for healthcare decision support systems with high accuracy, F1 score, and receiver operating characteristic area under the curve (ROC AUC), along with reduced training times. The results from this research offer important insights for enhancing decision-making strategies in the diagnosis and management of breast cancer.

A Model and Data Dual-Driven Framework for Lithium-Ion Battery Cycle Life Prediction Integrating Uncertainty Model

Abstract To address the shortcomings of single model-based or data-driven methods in battery life prediction, this paper proposes a model and data dual-driven framework for lithium-ion battery (LIB) cycle life prediction integrating uncertainty model. First, a cloud model is used to quantify the uncertainty inherent in the empirical model and construct a capacity degradation uncertainty model. Next, health features highly correlated with capacity are extracted from historical data to train a convolutional neural network, forming an online estimation model that guides subsequent parameter correction. Then, based on the prediction discrepancy between the two models, the error-feedback gate recurrent unit with improved attention model is developed for the online dynamic correction of empirical model parameters, realizing a closed-loop prediction framework. Finally, a rolling correction strategy based on statistical error is proposed to dynamically adjust the cloud model parameters, with the corrections fed back to the capacity degradation uncertainty model to further refine prediction uncertainty and accuracy. Experimental results indicate that the proposed method achieves high accuracy across various battery datasets, with relative error remaining below 2%, and provides uncertainty-quantified remaining useful life prediction, which effectively supports the online health monitoring and maintenance strategy formulation of LIBs in diverse scenarios.

Optimization of Stock Predictions on Indonesia Stock Exchange: A New Hybrid Deep Learning Method

This study presents a new method for predicting stock prices on the Indonesia Stock Exchange (IDX) using a hybrid deep learning model. The proposed model combines historical price data, consisting of open, high, low, and close values, with technical indicators such as Moving Average (MA), Simple Moving Average (SMA), and Exponential Moving Average (EMA). The proposed model offered improved accuracy and efficiency using distinct architectures of Recurrent Neural Network (RNN) and Gated Recurrent Unit (GRU) to process the datasets. The results showed that the proposed hybrid model significantly outperformed traditional single-architecture models in terms of R2, achieving 0.98 for the BBCA stock, surpassing models using only RNN or GRU. In addition, comparable improvements were observed with additional equities, such as the PT. Bank Mandiri Tbk (BMRI) and PT. Bank Negara Indonesia Tbk (BBNI) stocks, achieving an R2 of 0.99, demonstrating the proficiency of the proposed model in capturing the complex dynamics of the stock market. The results demonstrated the significant potential of combining historical data and technical indicators into the modeling procedure to predict stock prices. This process can benefit investors and economic forecasters in the stock market. The results could be further expanded by classifying datasets and investigating different sets of models to improve the performance of financial forecasting.

Comprehensive Learning Salp Swarm Algorithm with Ensemble Deep Learning-based ECG Signal Classification on Internet of Things Environment

The Internet of Things (IoT) in healthcare relates to implementing interconnected devices and systems for collecting and sharing healthcare information in real time. The integration of IoT in healthcare has the potential to enhance patient outcomes, reduce healthcare costs, and improve the efficacy of medical services. Electrocardiogram (ECG) is a non-invasive heart monitoring method that has become widely accessible due to user-friendly, low-cost, and lead-free wearable heart monitors. However, relying on overworked caregivers for manual monitoring is inefficient. This study develops a Comprehensive Learning Salp Swarm Algorithm with Ensemble Deep Learning (CLSSA-EDL) technique for ECG signal classification in IoT healthcare. The objective of CLSSA-EDL is to detect and classify ECG signals to support decision-making in the IoT healthcare environment. The CLSSA-EDL approach employs the DenseNet201 feature extraction method, with hyperparameters optimally selected by the CLSSA system. For ECG signal detection and classification, an ensemble model using a Stacked Autoencoder (SAE), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) is utilized. The CLSSA-EDL technique was evaluated on a benchmark ECG dataset, achieving an accuracy of 98.7%, sensitivity of 97.5%, and specificity of 99.1%, demonstrating superior performance compared to recent algorithms.

Integrated U-Net segmentation and gated recurrent unit classification for accurate brain tumor diagnosis from magnetic resonance imaging images

Early diagnosis and proper grouping of tumors in the brain are critical for successful therapy and positive outcomes for patients. This work proposes a complete technique for identifying brain tumors that employ sophisticated artificial intelligence methodologies and achieve an accuracy rate of 97.18%. The work makes use of the brain tumor magnetic resonance imaging (MRI) collection in Kaggle, which has 723 MRI scans classified as glioma, meningioma, pituitary tumor, and no tumor. These images are initially preprocessed, which includes scaling to a homogeneous size normalizing, and removal of noise to ensure uniformity and clarity. To improve the information set, generative adversarial networks (GANs) are used to perform data augmentation, producing artificial pictures that improve the database variety and resilience. To achieve exact cancer localization, the U-Net construction, recognized for its encoder-decoder design and skip links, is used to divide up tumor areas across images generated by MRI. The image segments are then input into gated recurrent units (GRUs), to analyze a collection of features to capture periods and differences between segments. The last classification is accomplished using an entirely linked layer and then a softmax stimulation, which provides the tumors classes. This method helps for medical experiments and clinical methods.

Seizure Detection in Medical IoT: Hybrid CNN-LSTM-GRU Model with Data Balancing and XAI Integration

The brain acts as the body’s central command, overseeing diverse functions including thought, memory, speech, movement, and the regulation of various organs. When healthy, the brain functions seamlessly and automatically; however, disruptions can lead to serious conditions such as Alzheimer’s Disease, Brain Cancer, Stroke, and Epilepsy. Epilepsy, a neurological disorder marked by recurrent seizures, results from irregular electrical activity in the brain. These seizures, which can strain both patients and neurologists, are characterized by symptoms like the loss of awareness, unusual behavior, and confusion. This study presents an efficient EEG-based epileptic seizure detection framework utilizing a hybrid Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) models approach to support automated and accurate diagnosis. Handling imbalanced EEG data, which can otherwise bias model outcomes and reduce predictive accuracy, is a key focus. Experimental results indicate that the proposed framework generally outperforms other Deep Learning and Machine Learning techniques with the highest accuracy at 99.13%. Likewise, an Explainable Artificial Intelligence (XAI) called SHAP (SHapley Additive exPlanations) is utilized to analyze the results and to improve the interpretability of the models from medical decision-making. This framework aligns with the objectives of the Medical Internet of Things (MIoT), advancing smart medical applications and services for effective epileptic seizure detection.

State of Health Estimation of Lithium-Ion Batteries Based on Hybrid Neural Networks with Residual Connections

Accurately estimating the state of health (SOH) of lithium-ion batteries is essential for ensuring the stability and safety of the battery. Although the hybrid neural network model demonstrates strong performance in estimating the SOH, network degradation becomes a significant issue as the depth of the neural network increases, potentially undermining the accuracy of the estimation. This paper presents a hybrid neural network estimation model with residual connections to address the issue of network degradation. First, the model utilizes a combination of convolutional neural networks and an attention mechanism to automatically extract feature information highly correlated with SOH from the partial charging data of lithium-ion batteries. Subsequently, a multi-layer gated recurrent unit (GRU) is employed to capture temporal information within the extracted features. To address the issue of network degradation that arises from stacking multiple layers of neural networks, residual connections are incorporated into the multi-layer GRUs, mitigating the accumulation of errors within deep networks. Finally, three distinct datasets are employed to validate the proposed model. The experimental results demonstrate that the model exhibits an average mean absolute error and root mean square error of less than 1.8% on both of these datasets.

Application of correlation-based recurrent neural network in porosity prediction for petroleum exploration

Abstract Porosity prediction serves as a key metric to determine the characteristics of reservoirs and to reduce the exploration efforts, which in turn lead to enhancing the production planning and making economic vital decisions. In this sense, one can use low costly and low time-consumptive method by conducting experimental tests on various soil samples. Predicting porosity can be effectively utilized for identifying suitable drilling locations for petroleum exploration, thus reducing the need for extensive laboratory experiments beforehand. The seismic behavior has a direct impact on porosity features. This study works to find a mapping relationship between well-log porosity and seismic attributes. A Pearson correlation coefficient method has been proposed based on diverse scales to determine the most significant seismic attributes and to reduce data dimensionality. The recurrent prediction neural network models have been proposed for learning and prediction. Three neural network models are introduced, represented by Nonlinear Autoregressive (NARX) network, the Long Short-Term Memory (LSTM) network, and the Gated Recurrent Unit (GRU) network. All these proposed models are trained to associate the nonlinear relationship between the selected attributes and porosity. The seismic attributes are assigned as the inputs to the neural models, while the porosity represents the target of neural structure. Moreover, the correlated and uncorrelated strategies are involved to test the effective attributes based on seven scenarios with the proposed neural structures. The results showed that the correlated NARX network consistently delivers better prediction accuracy and minimum errors as compared to other models for different correlation scales. Consequently, correlated NARX network is the best choice for porosity prediction.

MULTI-CLASS AUTOMATED SPEECH LANGUAGE RECOGNITION USING NATURAL LANGUAGE PROCESSING WITH OPTIMAL DEEP LEARNING MODEL

With technological development, human–computer interaction (HCI) has improved, and spoken communication among machines and humans is one solution to enhance and expedite this process. Researchers have recently explored several systems to improve speech and speaker recognition performance in recent decades. A crucial threat in HCI is developing models that can effectually listen and respond like humans. It resulted in the development of the automated speech emotion recognition (SER) method, which can recognize various emotional classes by electing and extracting effectual features from speech signals. The fundamental problem of automated speech detection is the considerable variation in speech signals because of distinct speakers, language differences, speech differences, contents and acoustic conditions, voice modulation differences based on age and gender. With enhancements in deep learning (DL) and the affordability of computational resources, specifically graphical processing units (GPUs), research underwent a paradigm shift. Therefore, this study develops a multi-class automated speech language recognition using natural language processing with optimal deep learning (MASLR-NLPODL) technique. The MASLR-NLPODL technique intends to accomplish the efficient identification of different spoken languages. In the MASLR-NLPODL technique, the initial preprocessing technique involves windowing, frame blocking, and pre-emphasis block. Next, an adaptive time-frequency feature extractor approach utilizing the discrete fractional Fourier transform (DFrFT) was applied, which can be attained by extending the discrete Fourier transform (DFT) with eigenvectors. An improved Harris hawks optimization (IHHO) technique can be employed to select effectual features. Moreover, the classification of spoken languages can be performed by the gated recurrent unit (GRU) model. Finally, the salp swarm algorithm (SSA)-based hyperparameter selection process is involved in enhancing the performance of the GRU model. The design of the IHHO-based feature selection and SSA-based hyperparameter tuning process demonstrates the novelty of the work. The performance evaluation of the MASLR-NLPODL technique takes place under the VoxForge Dataset. The experimental validation of the MASLR-NLPODL technique exhibited a superior accuracy outcome of 96.40% over existing techniques.

Integrating deep learning algorithms for forecasting evapotranspiration and assessing crop water stress in agricultural water management.

The increasing impacts of climate change on global agriculture necessitate the development of advanced predictive models for efficient water management in crop fields. This study aims to enhance the forecasting of evapotranspiration (ET), potential evapotranspiration (PET), and crop water stress index (CWSI) using state-of-the-art deep learning techniques. This research integrates high-resolution climatic data from the ACCESS-ESM model and incorporates four shared socioeconomic pathways (SSPs) to represent a wide range of future climate scenarios. We employ feed forward neural networks (FFNNs), convolutional neural networks (CNNs), gated recurrent units (GRUs), and long short-term memory networks (LSTMs) to predict ET, PET, and CWSI. These findings reveal significant improvements in prediction accuracy, offering valuable insights for agricultural water management in Bangladesh. This approach provides a robust framework for optimizing irrigation practices and enhancing crop resilience against climate-induced water stress.

Synergistic effect of artificial intelligence and new real-time disassembly sensors: Overcoming limitations and expanding application scope

Abstract With the acceleration of industrialization and the increase in the complexity of equipment, it is particularly important to accurately predict and effectively maintain equipment failures. In response to the problem of difficulty in accurately achieving predictive maintenance and the severely limited application range of new real-time disassembly sensors, this article conducts research on the synergistic effect of the new real-time disassembly sensors and artificial intelligence technology. First, relevant parameter data of a certain enterprise’s generator can be collected on-site, and the Pearson correlation coefficient method can be used to calculate the correlation between the generator data parameters and the target fault type, ensuring the degree of correlation in extracting data features. Then, based on the gated recurrent unit (GRU) model, the article applied the particle swarm optimization (PSO) algorithm to optimize the parameters of the GRU network and used the support vector machine (SVM) model to optimize the classification function of the network output. Finally, the optimized GRU model can be applied to predict the fault types of generators, and based on this, it can be applied to the energy industry, agriculture, medical and health, and transportation industries to verify the application scalability of artificial intelligence and new real-time disassembly sensors. The experimental results show that the optimized GRU model combined with the new real-time disassembly sensor achieved an average accuracy of 0.94 in generator fault prediction, with a cost loss rate of only 3.03%, a decrease of 5.91% compared to the single new real-time disassembly sensor. The combination of artificial intelligence technology for precise predictive maintenance of generators greatly reduces maintenance costs and overcomes the limitations of individual real-time sensor disassembly. This has to some extent expanded the application scope and promoted the intelligent development of various industries.

PilotCareTrans Net: an EEG data-driven transformer for pilot health monitoring

IntroductionIn high-stakes environments such as aviation, monitoring cognitive, and mental health is crucial, with electroencephalogram (EEG) data emerging as a keytool for this purpose. However traditional methods like linear models Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) architectures often struggle to capture the complex, non-linear temporal dependencies in EEG signals. These approaches typically fail to integrate multi-scale features effectively, resulting in suboptimal health intervention decisions, especially in dynamic, high-pressure environments like pilot training.MethodsTo overcome these challenges, this study introduces PilotCareTrans Net, a novel Transformer-based model designed for health intervention decision-making in aviation students. The model incorporates dynamic attention mechanisms, temporal convolutional layers, and multi-scale feature integration, enabling it to capture intricate temporal dynamics in EEG data more effectively. PilotCareTrans Net was evaluated on multiple public EEG datasets, including MODA, STEW, SJTUEmotion EEG, and Sleep-EDF, where it outperformed state-of-the-art models in key metrics.Results and discussionThe experimental results demonstrate the model's ability to not only enhance prediction accuracy but also reduce computational complexity, making it suitable for real-time applications in resource-constrained settings. These findings indicate that PilotCareTrans Net holds significant potential for improving cognitive health monitoring and intervention strategies in aviation, thereby contributing to enhanced safety and performance in critical environments.

A structural health monitoring data reconstruction method based on VMD and SSA-optimized GRU model

In the field of Structural Health Monitoring (SHM), complete datasets are fundamental for modal identification analysis and risk prediction. However, data loss due to sensor failures, transmission interruptions, or hardware issues is a common problem. To address this challenge, this study develops a method combining Variational Mode Decomposition (VMD) and Sparrow Search Algorithm (SSA)-optimized Gate Recurrent Unit (GRU) for recovering structural response data. The methodology initially employs Variational Mode Decomposition (VMD) to preprocess and decompose the existing data from the target sensor into Intrinsic Mode Functions (IMFs) and residuals. Subsequently, the Gated Recurrent Unit (GRU) network utilizes data from other sensors to reconstruct the IMFs and residuals, ultimately producing the data reconstruction results. During this process, Singular Spectrum Analysis (SSA) is used to optimize the hyperparameters of the GRU network. To validate the effectiveness of this method, we utilized one month of monitoring data collected from a certain project and a publicly available dataset. On the public dataset, we tested performance at different data loss rates. Results show that, compared to a standalone GRU model and a VMD + GRU model, the VMD + SSA + GRU model’s reconstruction data root mean squared error is reduced by 46.61% and 32.57% on average, respectively, while the coefficient of determination increases by 38.74% and 18.50%. The data reconstruction method proposed in this study can accurately capture trends in missing data, without the need for manual hyperparameter tuning, and the reconstruction results are highly consistent with the real data.

HERGAT: predicting hERG blockers using graph attention mechanism through atom- and molecule-level interaction analyses

The human ether-a-go-go-related gene (hERG) channel plays a critical role in the electrical activity of the heart, and its blockers can cause serious cardiotoxic effects. Thus, screening for hERG channel blockers is a crucial step in the drug development process. Many in silico models have been developed to predict hERG blockers, which can efficiently save time and resources. However, previous methods have found it hard to achieve high performance and to interpret the predictive results. To overcome these challenges, we have proposed hERGAT, a graph neural network model with an attention mechanism, to consider compound interactions on atomic and molecular levels. In the atom-level interaction analysis, we applied a graph attention mechanism (GAT) that integrates information from neighboring nodes and their extended connections. The hERGAT employs a gated recurrent unit (GRU) with the GAT to learn information between more distant atoms. To confirm this, we performed clustering analysis and visualized a correlation heatmap, verifying the interactions between distant atoms were considered during the training process. In the molecule-level interaction analysis, the attention mechanism enables the target node to focus on the most relevant information, highlighting the molecular substructures that play crucial roles in predicting hERG blockers. Through a literature review, we confirmed that highlighted substructures have a significant role in determining the chemical and biological characteristics related to hERG activity. Furthermore, we integrated physicochemical properties into our hERGAT model to improve the performance. Our model achieved an area under the receiver operating characteristic of 0.907 and an area under the precision-recall of 0.904, demonstrating its effectiveness in modeling hERG activity and offering a reliable framework for optimizing drug safety in early development stages.Scientific contribution:hERGAT is a deep learning model for predicting hERG blockers by combining GAT and GRU, enabling it to capture complex interactions at atomic and molecular levels. We improve the model's interpretability by analyzing the highlighted molecular substructures, providing valuable insights into their roles in determining hERG activity. The model achieves high predictive performance, confirming its potential as a preliminary tool for early cardiotoxicity assessment and enhancing the reliability of the results.

Sustainability, Accuracy, Fairness, and Explainability (SAFE) Machine Learning in Quantitative Trading

The paper investigates the application of advanced machine learning (ML) methodologies, with a particular emphasis on state-of-the-art deep learning models, to predict financial market dynamics and maximize profitability through algorithmic trading strategies. The study compares the predictive capabilities and behavioral characteristics of traditional machine learning approaches, such as logistic regression and support vector machines, with those of highly sophisticated deep learning architectures, including Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and Gated Recurrent Units (GRUs). The findings underscore the fundamental distinctions between these methodologies, with deeply trained models exhibiting markedly different predictive behaviors and performance, particularly in capturing complex temporal patterns within financial data. A cornerstone of the paper is the introduction and rigorous analysis of a framework to evaluate models, by means of the SAFE framework (Sustainability, Accuracy, Fairness, and Explainability). The framework is designed to address the opacity of black-box ML models by systematically evaluating their behavior across a set of critical dimensions. It also demonstrates how models’ predictive outputs align with the observed data, thereby reinforcing their reliability and robustness. The paper leverages historical stock price data from International Business Machines Corporation (IBM). The dataset is partitioned into a training phase during which the models are calibrated, and a validation phase, used to evaluate the predictive performance of the generated trading signals. The study addresses two primary machine learning tasks: regression and classification. Classical models are utilized for classification tasks, with their outputs directly interpreted as trading signals, while advanced deep learning models are employed for regression, with predictions of future stock prices further processed into actionable trading strategies. To evaluate the effectiveness of each strategy, rigorous backtesting is conducted, incorporating visual representations such as equity curves to assess profitability and key risk metrics like maximum drawdown for risk management. Supplementary performance indicators, including hit rates and the incidence of false positions, are analyzed alongside the equity curves to provide a holistic assessment of each model’s performance. This comprehensive evaluation not only highlights the superiority of cutting-edge deep learning models in predicting financial market trends but also demonstrates the pivotal role of the SAFE framework in ensuring that machine learning models remain trustworthy, interpretable, and aligned with ethical considerations.

An Optimized Weighted-Voting-Based Ensemble Learning Approach for Fake News Classification

The emergence of diverse content-sharing platforms and social media has rendered the dissemination of fake news and misinformation increasingly widespread. This misinformation can cause extensive confusion and fear throughout the populace. Confronting this dilemma necessitates an effective and accurate approach to identifying misinformation, an intrinsically intricate process. This research introduces an automated and efficient method for detecting false information. We evaluated the efficacy of various machine learning and deep learning models on two separate fake news datasets of differing sizes via holdout cross-validation. Furthermore, we evaluated the efficacy of three distinct word vectorization methods. Additionally, we employed an enhanced weighted voting ensemble model that enhances fake news detection by integrating logistic regression (LR), support vector machine (SVM), gated recurrent unit (GRU), and long short-term memory (LSTM) networks. This method exhibits enhanced performance relative to previous techniques: 98.76% for the PolitiFact dataset and 97.67% for the BuzzFeed dataset. Furthermore, the model outperforms individual components, resulting in superior accuracy, precision, recall, and F1 scores. The enhancements in performance result from the ensemble method’s capacity to use the advantages of each base model, hence providing robust generalization across datasets. Cross-validation was employed to enhance the model’s trustworthiness, validating its capacity to generalize effectively to novel data.

A machine learning technique for optimizing load demand prediction within air conditioning systems utilizing GRU/IASO model

Air conditioning systems are widely used to provide thermal comfort in hot and humid regions, but they also consume a large amount of energy. Therefore, accurate and reliable load demand forecasting is essential for energy management and optimization in air conditioning systems. Within the current paper, a novel model on the basis of machine learning has been presented for dynamic optimal load demand forecasting in air conditioning systems. The model is based on using an optimized design of Gated recurrent unit (GRU) network and an enhanced metaheuristic algorithm, named Improved Alpine Skiing Optimizer (IASO). GRU is a recurrent neural network that has the ability to comprehend intricate temporal relationships within the input data. On the other hand, the IASO technique has been considered to be a population-based optimization technique emulating the downhill skiing behavior of skiers. The proposed GRU/IASO model is trained and tested utilizing data of real-world obtained through a commercial complex situated within an area characterized by high humidity and hot climate. By comparing the proposed method with some other commonly used techniques, including ---, the advantage of the suggested model regarding accuracy and robustness has been defined.