Conventional Machine Learning Models Research Articles

A novel ensemble model for detecting fake news

<div>Due the growing proliferation of fake news over the past couple of years, our</div><div>objective in this paper is to propose an ensemble model for the automatic classification of article news as being either real or fake. For this purpose, we opt</div><div>for a blending technique that combines three models, namely bidirectional long</div><div>short-term memory (Bi-LSTM), stochastic gradient descent classifier and ridge</div><div>classifier. The implementation of the proposed model (i.e. BI-LSR) on real</div><div>world datasets, has shown outstanding results. In fact, it achieved an accuracy</div><div>score of 99.16%. Accordingly, this ensemble learning has proven to do perform</div><div>better than individual conventional machine learning and deep learning models</div><div>as well as many ensemble learning approaches cited in the literature.</div>

Speaker identification under noisy conditions using hybrid convolutional neural network and gated recurrent unit

<p><span>Speaker identification is biometrics that classifies or identifies a person from other speakers based on speech characteristics. Recently, deep learning models outperformed conventional machine learning models in speaker identification. Spectrograms of the speech have been used as input in deep learning-based speaker identification using clean speech. However, the performance of speaker identification systems gets degraded under noisy conditions. Cochleograms have shown better results than spectrograms in deep learning-based speaker recognition under noisy and mismatched conditions. Moreover, hybrid convolutional neural network (CNN) and recurrent neural network (RNN) variants have shown better performance than CNN or RNN variants in recent studies. However, there is no attempt conducted to use a hybrid CNN and enhanced RNN variants in speaker identification using cochleogram input to enhance the performance under noisy and mismatched conditions. In this study, a speaker identification using hybrid CNN and the gated recurrent unit (GRU) is proposed for noisy conditions using cochleogram input. VoxCeleb1 audio dataset with real-world noises, white Gaussian noises (WGN) and without additive noises were employed for experiments. The experiment results and the comparison with existing works show that the proposed model performs better than other models in this study and existing works</span><span>.</span></p>

A Hierarchical Deep Learning Architecture for Robust Intrusion Detection in Time-Evolving Security Landscapes

Abstract: This paper introduces a novel architecture for Intrusion Detection Systems (IDS), designed to enhance resilience against adversarial attacks by integrating conventional machine learning (ML) models with Deep Learning (DL) models. The proposed system, termed DLL-IDS, comprises three key components: a DL-based IDS, an adversarial example (AE) detector based on local intrinsic dimensionality (LID), and an ML-based IDS. Initially, a novel AE detector is developed using LID to identify potential adversarial examples. This detector serves as a crucial component in identifying suspicious inputs that may be crafted to deceive the IDS. Subsequently, the system leverages the low transferability of attacks between DL and ML models to establish a robust ML model capable of discerning the maliciousness of potential AEs. The DLL-IDS architecture operates as follows: if incoming traffic is flagged as an AE by the detector, the MLbased IDS is employed to evaluate its maliciousness; otherwise, the DL-based IDS handles the prediction. By integrating both DL and ML approaches, the system benefits from the high prediction accuracy of DL models while exploiting the low susceptibility of ML models to adversarial attacks. Experimental results demonstrate significant enhancements in IDS prediction performance under adversarial conditions, achieving high accuracy with minimal resource consumption. This fusion mechanism effectively combines the strengths of DL and ML models, thereby bolstering the overall robustness of the IDS against sophisticated intrusion attempts.

Deep learning CNN-APSO-LSSVM hybrid fusion model for feature optimization and gas-bearing prediction

Conventional machine learning (CML) methods have been successfully applied for gas reservoir prediction. Their prediction accuracy largely depends on the quality of the sample data; therefore, feature optimization of the input samples is particularly important. Commonly used feature optimization methods increase the interpretability of gas reservoirs; however, their steps are cumbersome, and the selected features cannot sufficiently guide CML models to mine the intrinsic features of sample data efficiently. In contrast to CML methods, deep learning (DL) methods can directly extract the important features of targets from raw data. Therefore, this study proposes a feature optimization and gas-bearing prediction method based on a hybrid fusion model that combines a convolutional neural network (CNN) and an adaptive particle swarm optimization-least squares support vector machine (APSO-LSSVM). This model adopts an end-to-end algorithm structure to directly extract features from sensitive multicomponent seismic attributes, considerably simplifying the feature optimization. A CNN was used for feature optimization to highlight sensitive gas reservoir information. APSO-LSSVM was used to fully learn the relationship between the features extracted by the CNN to obtain the prediction results. The constructed hybrid fusion model improves gas-bearing prediction accuracy through two processes of feature optimization and intelligent prediction, giving full play to the advantages of DL and CML methods. The prediction results obtained are better than those of a single CNN model or APSO-LSSVM model. In the feature optimization process of multicomponent seismic attribute data, CNN has demonstrated better gas reservoir feature extraction capabilities than commonly used attribute optimization methods. In the prediction process, the APSO-LSSVM model can learn the gas reservoir characteristics better than the LSSVM model and has a higher prediction accuracy. The constructed CNN-APSO-LSSVM model had lower errors and a better fit on the test dataset than the other individual models. This method proves the effectiveness of DL technology for the feature extraction of gas reservoirs and provides a feasible way to combine DL and CML technologies to predict gas reservoirs.

PlantNh-Kcr: a deep learning model for predicting non-histone crotonylation sites in plants

BackgroundLysine 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.ResultsMost 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.ConclusionPlantNh-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.

Prediction of emergency department revisits among child and youth mental health outpatients using deep learning techniques

BackgroundThe proportion of Canadian youth seeking mental health support from an emergency department (ED) has risen in recent years. As EDs typically address urgent mental health crises, revisiting an ED may represent unmet mental health needs. Accurate ED revisit prediction could aid early intervention and ensure efficient healthcare resource allocation. We examine the potential increased accuracy and performance of graph neural network (GNN) machine learning models compared to recurrent neural network (RNN), and baseline conventional machine learning and regression models for predicting ED revisit in electronic health record (EHR) data.MethodsThis study used EHR data for children and youth aged 4–17 seeking services at McMaster Children’s Hospital’s Child and Youth Mental Health Program outpatient service to develop and evaluate GNN and RNN models to predict whether a child/youth with an ED visit had an ED revisit within 30 days. GNN and RNN models were developed and compared against conventional baseline models. Model performance for GNN, RNN, XGBoost, decision tree and logistic regression models was evaluated using F1 scores.ResultsThe GNN model outperformed the RNN model by an F1-score increase of 0.0511 and the best performing conventional machine learning model by an F1-score increase of 0.0470. Precision, recall, receiver operating characteristic (ROC) curves, and positive and negative predictive values showed that the GNN model performed the best, and the RNN model performed similarly to the XGBoost model. Performance increases were most noticeable for recall and negative predictive value than for precision and positive predictive value.ConclusionsThis study demonstrates the improved accuracy and potential utility of GNN models in predicting ED revisits among children and youth, although model performance may not be sufficient for clinical implementation. Given the improvements in recall and negative predictive value, GNN models should be further explored to develop algorithms that can inform clinical decision-making in ways that facilitate targeted interventions, optimize resource allocation, and improve outcomes for children and youth.

Identification and Classification of Coix seed Storage Years Based on Hyperspectral Imaging Technology Combined with Deep Learning

Developing a fast and non-destructive methodology to identify the storage years of Coix seed is important in safeguarding consumer well-being. This study employed the utilization of hyperspectral imaging (HSI) in conjunction with conventional machine learning techniques such as support vector machines (SVM), k-nearest neighbors (KNN), random forest (RF), extreme gradient boosting (XGBoost), as well as the deep learning method of residual neural network (ResNet), to establish identification models for Coix seed samples from different storage years. Under the fusion-based modeling approach, the model’s classification accuracy surpasses that of visible to near infrared (VNIR) and short-wave infrared (SWIR) spectral modeling individually. The classification accuracy of the ResNet model and SVM exceeds that of other conventional machine learning models (KNN, RF, and XGBoost). Redundant variables were further diminished through competitive adaptive reweighted sampling feature wavelength screening, which had less impact on the model’s accuracy. Upon validating the model’s performance using an external validation set, the ResNet model yielded more satisfactory outcomes, exhibiting recognition accuracy exceeding 85%. In conclusion, the comprehensive results demonstrate that the integration of deep learning with HSI techniques effectively distinguishes Coix seed samples from different storage years.

Predicting Hypoxia Using Machine Learning: Systematic Review.

Hypoxia is an important risk factor and indicator for the declining health of inpatients. Predicting future hypoxic events using machine learning is a prospective area of study to facilitate time-critical interventions to counter patient health deterioration. This systematic review aims to summarize and compare previous efforts to predict hypoxic events in the hospital setting using machine learning with respect to their methodology, predictive performance, and assessed population. A systematic literature search was performed using Web of Science, Ovid with Embase and MEDLINE, and Google Scholar. Studies that investigated hypoxia or hypoxemia of hospitalized patients using machine learning models were considered. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. After screening, a total of 12 papers were eligible for analysis, from which 32 models were extracted. The included studies showed a variety of population, methodology, and outcome definition. Comparability was further limited due to unclear or high risk of bias for most studies (10/12, 83%). The overall predictive performance ranged from moderate to high. Based on classification metrics, deep learning models performed similar to or outperformed conventional machine learning models within the same studies. Models using only prior peripheral oxygen saturation as a clinical variable showed better performance than models based on multiple variables, with most of these studies (2/3, 67%) using a long short-term memory algorithm. Machine learning models provide the potential to accurately predict the occurrence of hypoxic events based on retrospective data. The heterogeneity of the studies and limited generalizability of their results highlight the need for further validation studies to assess their predictive performance.

Atomistic simulation assisted error-inclusive Bayesian machine learning for probabilistically unraveling the mechanical properties of solidified metals

Solidification phenomenon has been an integral part of the manufacturing processes of metals, where the quantification of stochastic variations and manufacturing uncertainties is critically important. Accurate molecular dynamics (MD) simulations of metal solidification and the resulting properties require excessive computational expenses for probabilistic stochastic analyses where thousands of random realizations are necessary. The adoption of inadequate model sizes and time scales in MD simulations leads to inaccuracies in each random realization, causing a large cumulative statistical error in the probabilistic results obtained through Monte Carlo (MC) simulations. In this work, we present a machine learning (ML) approach, as a data-driven surrogate to MD simulations, which only needs a few MD simulations. This efficient yet high-fidelity ML approach enables MC simulations for full-scale probabilistic characterization of solidified metal properties considering stochasticity in influencing factors like temperature and strain rate. Unlike conventional ML models, the proposed hybrid polynomial correlated function expansion here, being a Bayesian ML approach, is data efficient. Further, it can account for the effect of uncertainty in training data by exploiting mean and standard deviation of the MD simulations, which in principle addresses the issue of repeatability in stochastic simulations with low variance. Stochastic numerical results for solidified aluminum are presented here based on complete probabilistic uncertainty quantification of mechanical properties like Young’s modulus, yield strength and ultimate strength, illustrating that the proposed error-inclusive data-driven framework can reasonably predict the properties with a significant level of computational efficiency.

Multi-modal LSTM network for anomaly prediction in piston engine aircraft

An aircraft is a highly intricate system that features numerous subsystems, assemblies, and individual components for which regular maintenance is inevitable. The operational efficiency of an aircraft can be maximised, and its maintenance needs can be reduced using an effective yet automatic AI-based health monitoring systems which are more efficient as compared to designing and constructing expensive and harder to operate engine testbeds. It has been observed that aircraft engine anomalies such as undergoing flameouts can occur due to the rapid change in the temperature of the engine. Engine oil temperature and cylinder head temperature, two measures connected to this issue, might be affected differently depending on flight modes and operational conditions which in turn hamper AI-based algorithms to yield accurate prediction on engine failures. In general, previous studies lack comprehensive analysis on anomaly prediction in piston engine aircraft using modern machine learning solutions. Furthermore, abrupt variation in aircraft sensors' data and noise result in either overfitting or unfavourable performance by such techniques. This work aims at studying conventional machine learning and deep learning models to foretell the possibility of engine flameout using engine oil and cylinder head temperatures of a widely used Textron Lycoming IO-540 six-cylinder piston engine. This is achieved through pre-processing the data extracted from the aircraft's real-time flight data recorder followed by prediction using specially designed multi-modal regularised Long Short-Term Memory network to enhance generalisation and avoid overfitting on highly variable data. The proposed architecture yields improved results with root mean square error of 0.55 and 3.20 on cylinder head and engine oil temperatures respectively averaged over three case studies of five different flights. These scores are significantly better i.e., up to 84% as compared to other popular machine learning predictive approaches including Random Forest, Decision Tree Regression, Artificial Neural Networks and vanilla Long Short-Term Memory networks. Through performance evaluation, it can be established that the proposed system is capable of predicting engine flameout 2 minutes ahead and is suitable for integration with the software system of aircraft's engine control unit.

A deep learning approach for health monitoring in rotating machineries using vibrations and thermal features

Gearbox failures can lead to substantial damage, significant financial losses due to maintenance downtimes, and, in some instances, fatalities. This study introduces an intelligent gear fault diagnosis system employing a convolutional neural network (CNN), utilizing vibration and thermal features extracted from healthy, chipped, and broken tooth gear health categories. The performance of the CNN is compared with conventional machine learning models, including Naïve Bayes (NB), random forest (RF), and support vector machine (SVM) classifiers. Experimental investigations highlight CNN’s remarkable performance. With vibration features, the CNN achieved 96.78% accuracy, surpassing SVM (84.89%), NB (81.56%), and RF (85.11%). The CNN attained 100% accuracy when utilizing thermal features, while SVM, NB, and RF achieved 91.11%, 88.89%, and 96.51% accuracies, respectively.

Hyperspectral Imaging Using a Convolutional Neural Network with Transformer for the Soluble Solid Content and pH Prediction of Cherry Tomatoes.

Cherry tomatoes are cultivated worldwide and favored by consumers of different ages. The soluble solid content (SSC) and pH are two of the most important quality attributes of cherry tomatoes. The rapid and non-destructive measurement of the SSC and pH of cherry tomatoes is of great significance to their production and consumption. In this research, hyperspectral imaging combined with a convolutional neural network with Transformer (CNN-Transformer) was utilized to analyze the SSC and pH of cherry tomatoes. Conventional machine learning and deep learning models were established for the determination of the SSC and pH. The findings demonstrated that CNN-Transformer yielded outstanding results in predicting the SSC, with the coefficient of determination of calibration (R2C), validation (R2V), and prediction (R2P) for the SSC being 0.83, 0.87, and 0.83, respectively. Relatively worse results were obtained for the pH value prediction, with R2C, R2V, and R2P values of 0.74, 0.68, and 0.60, respectively. Furthermore, the visualization of the CNN-Transformer model revealed the wavelength weight distributions, indicating that the 1380-1650 nm range served as the characteristic band for the SSC, while the spectral range at 945-1280 nm was the characteristic band for pH. In conclusion, integrating spectral information features with the attention mechanism of Transformer through a convolutional neural network can enhance the accuracy of predicting the SSC and pH for cherry tomatoes.

Enhancing crop recommendation systems with explainable artificial intelligence: a study on agricultural decision-making

Crop Recommendation Systems are invaluable tools for farmers, assisting them in making informed decisions about crop selection to optimize yields. These systems leverage a wealth of data, including soil characteristics, historical crop performance, and prevailing weather patterns, to provide personalized recommendations. In response to the growing demand for transparency and interpretability in agricultural decision-making, this study introduces XAI-CROP an innovative algorithm that harnesses eXplainable artificial intelligence (XAI) principles. The fundamental objective of XAI-CROP is to empower farmers with comprehensible insights into the recommendation process, surpassing the opaque nature of conventional machine learning models. The study rigorously compares XAI-CROP with prominent machine learning models, including Gradient Boosting (GB), Decision Tree (DT), Random Forest (RF), Gaussian Naïve Bayes (GNB), and Multimodal Naïve Bayes (MNB). Performance evaluation employs three essential metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R2). The empirical results unequivocally establish the superior performance of XAI-CROP. It achieves an impressively low MSE of 0.9412, indicating highly accurate crop yield predictions. Moreover, with an MAE of 0.9874, XAI-CROP consistently maintains errors below the critical threshold of 1, reinforcing its reliability. The robust R2 value of 0.94152 underscores XAI-CROP's ability to explain 94.15% of the data's variability, highlighting its interpretability and explanatory power.

Extrapolative prediction of small-data molecular property using quantum mechanics-assisted machine learning

Data-driven materials science has realized a new paradigm by integrating materials domain knowledge and machine-learning (ML) techniques. However, ML-based research has often overlooked the inherent limitation in predicting unknown data: extrapolative performance, especially when dealing with small-scale experimental datasets. Here, we present a comprehensive benchmark for assessing extrapolative performance across 12 organic molecular properties. Our large-scale benchmark reveals that conventional ML models exhibit remarkable performance degradation beyond the training distribution of property range and molecular structures, particularly for small-data properties. To address this challenge, we introduce a quantum-mechanical (QM) descriptor dataset, called QMex, and an interactive linear regression (ILR), which incorporates interaction terms between QM descriptors and categorical information pertaining to molecular structures. The QMex-based ILR achieved state-of-the-art extrapolative performance while preserving its interpretability. Our benchmark results, QMex dataset, and proposed model serve as valuable assets for improving extrapolative predictions with small experimental datasets and for the discovery of novel materials/molecules that surpass existing candidates.

Prediction of postoperative infectious complications in elderly patients with colorectal cancer: a study based on improved machine learning

BackgroundInfectious complications after colorectal cancer (CRC) surgery increase perioperative mortality and are significantly associated with poor prognosis. We aimed to develop a model for predicting infectious complications after colorectal cancer surgery in elderly patients based on improved machine learning (ML) using inflammatory and nutritional indicators.MethodsThe data of 512 elderly patients with colorectal cancer in the Third Affiliated Hospital of Anhui Medical University from March 2018 to April 2022 were retrospectively collected and randomly divided into a training set and validation set. The optimal cutoff values of NLR (3.80), PLR (238.50), PNI (48.48), LCR (0.52), and LMR (2.46) were determined by receiver operating characteristic (ROC) curve; Six conventional machine learning models were constructed using patient data in the training set: Linear Regression, Random Forest, Support Vector Machine (SVM), BP Neural Network (BP), Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBoost) and an improved moderately greedy XGBoost (MGA-XGBoost) model. The performance of the seven models was evaluated by area under the receiver operator characteristic curve, accuracy (ACC), precision, recall, and F1-score of the validation set.ResultsFive hundred twelve cases were included in this study; 125 cases (24%) had postoperative infectious complications. Postoperative infectious complications were notably associated with 10 items features: American Society of Anesthesiologists scores (ASA), operation time, diabetes, presence of stomy, tumor location, NLR, PLR, PNI, LCR, and LMR. MGA-XGBoost reached the highest AUC (0.862) on the validation set, which was the best model for predicting postoperative infectious complications in elderly patients with colorectal cancer. Among the importance of the internal characteristics of the model, LCR accounted for the highest proportion. Conclusions: This study demonstrates for the first time that the MGA-XGBoost model with 10 risk factors might predict postoperative infectious complications in elderly CRC patients.

Data-driven transfer learning framework for estimating on-ramp and off-ramp traffic flows

To develop the most appropriate control strategy and monitor, maintain, and evaluate the traffic performance of the freeway weaving areas, state and local Departments of Transportation need to have access to traffic flows at each pair of on-ramp and off-ramp. However, ramp flows are not always readily available to transportation agencies, and little effort has been made to estimate these missing traffic flows in locations where no physical sensors are installed. To bridge this research gap, a data-driven framework is proposed that can accurately estimate the missing ramp flows by solely using data collected from loop detectors on freeway mainlines. The proposed framework employs a transfer learning model. The transfer learning model relaxes the assumption that the underlying data distributions of the source and target domains must be the same. Therefore, the proposed framework can guarantee high-accuracy estimation of on-ramp and off-ramp flows on freeways with different traffic patterns, distributions, and characteristics. Based on the experimental results, the flow estimation mean absolute errors range between 23.90 veh/h to 40.85 veh/h for on-ramps and 31.58 veh/h to 45.31 veh/h for off-ramps; the flow estimation root mean square errors range between 34.55 veh/h to 57.77 veh/h for on-ramps, and 41.75 veh/h to 58.80 veh/h for off-ramps. Further, the comparison analysis shows that the proposed framework outperforms other conventional machine learning models. The estimated ramp flows based on the proposed method can help transportation agencies to enhance the operations of their ramp control strategies for locations where physical sensors are not installed.

Modelling clogging dynamics in groundwater systems using multiscale homogenized physics informed neural network (MHPINN)

Groundwater systems, or also commonly known as large-scale closed and saturated porous media systems in general, are important sources of potable water, especially in arid countries. However, maintenance of groundwater systems is operationally expensive and time-consuming, hence requiring field engineers to optimally predict the system’s breakthrough phase. To address this engineering challenge, this study developed an alternative physics informed machine learning model, termed as multiscale homogenized physics informed neural network (MHPINN), to model and predict the effluent discharge concentrations from closed and saturated porous media systems when removing different contaminants. The proposed MHPINN model mainly exploits the homogenization theory, coupled with multiscale perturbation analysis, for hydrodynamic modelling to extract important model input features pertaining to the pore-scale clogging behavior in porous media. Together with available experimental data, the extracted model features then set the input layer for training neural network model to predict the discharge concentration of emulated porous media groundwater systems. At the same time, the model training is performed with modified mean square error (MSE) cost function, based upon homogenized mathematical representations derived from a series of rigorous mathematical constructs and formulations, to better facilitate the model’s learning, especially under sparse data constraint. To verify the proposed methodology, MPINN is trained (and validated) with data obtained from five historical experimental studies, which emulated porous media systems to remove ammonium ions and fine-scaled colloids, to model and predict the discharge concentration in each study. Overall, MHPINN outperformed other conventional machine learning models (random forest, support vector machines, etc.) by obtaining higher predictive accuracy of around 4% on average from the model’s validation step across the five different experimental studies. Therefore, the predictive analysis underscored the effectiveness of MHPINN to combine both physics-based modelling and machine learning to model complex porous media systems, under sparse experimental data constraints, hence demonstrating the potential of the proposed methodology to extend to other physical systems.

Retinal Disease Classification Using Custom CNN Model From OCT Images

Retinal diseases affect the eye’s retina and are the leading cause of vision loss. Timely detection and diagnosis can provide more treatment options and preserve vision. Retinal imaging, such as Optical Coherence Tomography (OCT), is used for screening patients and enables clinicians to visualize the retinal layers in detail, helping in early diagnosis and tailoring appropriate treatment plans. This study uses OCT images to detect diseases related to macular disorders common in old age. A custom convolution Neural Network (CNN) model is proposed in this study to classify OCT images of normal, choroidal neovascularization (CNV), diabetic macular edema (DME), and Drusen from a public dataset. Experiments were conducted using conventional machine learning (ML) and transfer learning-based models other than the proposed one. The proposed model performed favorably, attaining the training and validation accuracy of 97% and 93%, respectively. The proposed model achieves a testing accuracy of 98% with a loss of 0.051 and outperforms the existing models compared to this study.

Traffic flow prediction with a multi-dimensional feature input: A new method based on attention mechanisms

<abstract> <p>Accurately predicting traffic flow is an essential component of intelligent transportation systems. The advancements in traffic data collection technology have broadened the range of features that affect and represent traffic flow variations. However, solely inputting gathered features into the model without analysis might overlook valuable information, hindering the improvement of predictive performance. Furthermore, intricate dynamic relationships among various feature inputs could constrain the model's potential for further enhancement in predictive accuracy. Consequently, extracting pertinent features from datasets and modeling their mutual influence is critical in attaining heightened precision in traffic flow predictions. First, we perform effective feature extraction by considering the temporal dimension and inherent operating rules of traffic flow, culminating in Multivariate Time Series (MTS) data used as input for the model. Then, an attention mechanism is proposed based on the MTS input data. This mechanism assists the model in selecting pertinent time series for multivariate forecasting, mitigating inter-feature influence, and achieving accurate predictions through the concentration on crucial information. Finally, empirical findings from real highway datasets illustrate the enhancement of predictive accuracy attributed to the proposed features within the model. In contrast to conventional machine learning or attention-based deep learning models, the proposed attention mechanism in this study demonstrates superior accuracy and stability in MTS-based traffic flow prediction tasks.</p> </abstract>

Functional Near‐Infrared Spectroscopy‐Based Computer‐Aided Diagnosis of Major Depressive Disorder Using Convolutional Neural Network with a New Channel Embedding Layer Considering Inter‐Hemispheric Asymmetry in Prefrontal Hemodynamic Responses

Background. Functional near‐infrared spectroscopy (fNIRS) is being extensively explored as a potential primary screening tool for major depressive disorder (MDD) because of its portability, cost‐effectiveness, and low susceptibility to motion artifacts. However, the fNIRS‐based computer‐aided diagnosis (CAD) of MDD using deep learning methods has rarely been studied. In this study, we propose a novel deep learning framework based on a convolutional neural network (CNN) for the fNIRS‐based CAD of MDD with high accuracy. Materials and Methods. The fNIRS data of participants—48 patients with MDD and 68 healthy controls (HCs)—were obtained while they performed a Stroop task. The hemodynamic responses calculated from the preprocessed fNIRS data were used as inputs to the proposed CNN model with an ensemble CNN architecture, comprising three 1D depth‐wise convolutional layers specifically designed to reflect interhemispheric asymmetry in hemodynamic responses between patients with MDD and HCs, which is known to be a distinct characteristic in previous MDD studies. The performance of the proposed model was evaluated using a leave‐one‐subject‐out cross‐validation strategy and compared with those of conventional machine learning and CNN models. Results. The proposed model exhibited a high accuracy, sensitivity, and specificity of 84.48%, 83.33%, and 85.29%, respectively. The accuracies of conventional machine learning algorithms—shrinkage linear discriminator analysis, regularized support vector machine, EEGNet, and ShallowConvNet—were 73.28%, 74.14%, 62.93%, and 62.07%, respectively. Conclusions. In conclusion, the proposed deep learning model can differentiate between the patients with MDD and HCs more accurately than the conventional models, demonstrating its applicability in fNIRS‐based CAD systems.