Deep Learning-based Neural Network Research Articles

Key Technologies for Autonomous Fruit- and Vegetable-Picking Robots: A Review

With the rapid pace of urbanization, a significant number of rural laborers are migrating to cities, leading to a severe shortage of agricultural labor. Consequently, the modernization of agriculture has become a priority. Autonomous picking robots represent a crucial component of agricultural technological innovation, and their development drives progress across the entire agricultural sector. This paper reviews the current state of research on fruit- and vegetable-picking robots, focusing on key aspects such as the vision system sensors, target detection, localization, and the design of end-effectors. Commonly used target recognition algorithms, including image segmentation and deep learning-based neural networks, are introduced. The challenges of target recognition and localization in complex environments, such as those caused by branch and leaf obstruction, fruit overlap, and oscillation in natural settings, are analyzed. Additionally, the characteristics of the three main types of end-effectors—clamping, suction, and cutting—are discussed, along with an analysis of the advantages and disadvantages of each design. The limitations of current agricultural picking robots are summarized, taking into account the complexity of operation, research and development costs, as well as the efficiency and speed of picking. Finally, the paper offers a perspective on the future of picking robots, addressing aspects such as environmental adaptability, functional diversity, innovation and technological convergence, as well as policy and farm management.

Machine Learning Approaches for Automated Diagnosis of Cardiovascular Diseases: A Review of Electrocardiogram Data Applications.

Cardiovascular diseases (CVDs) have been identified as the leading cause of mortality worldwide. Electrocardiogram (ECG) is a fundamental diagnostic tool used for the diagnosis and detection of these diseases. The new technological tools can help enhance the effectiveness of ECGs. Machine learning (ML) is widely acknowledged as a highly effective approach in the realm of computer-aided diagnostics. This article presents a review of the effectiveness of ML algorithms and deep-learning algorithms in diagnosing, identifying, and classifying CVDs using ECG data. The review identified relevant studies published in the 5 major databases: PubMed, Web of Science (WoS), Scopus, Springer, and IEEE Xplore; between 2021 and 2023, a total of 30 were chosen for the comprehensive quantitative and qualitative. The study demonstrated that different datasets are available online with data related to CVDs. The various ML techniques are employed for the purpose of classification. Based on our investigation, it has been observed that deep learning-based neural network algorithms, such as convolutional neural networks and deep neural networks, have demonstrated superior performance in the detection of entire record data. Furthermore, deep learning showcases its efficacy even when confronted with a scarcity of data. ML approaches utilizing ECG data exhibit a notable proficiency in the realm of diagnosis, hence holding the potential to mitigate the occurrence of disease-related consequences at advanced stages.

MobileNet-V2 /IFHO model for Accurate Detection of early-stage diabetic retinopathy

Diabetic retinopathy is a serious eye disease that may lead to loss of vision if it is not treated. Early detection is crucial in preventing further vision impairment and enabling timely interventions. Despite notable advancements in AI-based methods for detecting diabetic retinopathy, researchers are still striving to enhance the efficiency of these techniques. Therefore, obtaining an efficient technique in this field is essential. In this research, a new strategy has been proposed to improve the detection of diabetic retinopathy by increasing the accuracy of diagnosis and identifying cases in the initial stages. To achieve this, it has been proposed to integrate the MobileNet-V2 deep learning-based neural network with Improved Fire Hawk Optimizer (IFHO). The MobileNet-V2 network has been renowned for its efficiency and accuracy in image classification tasks, making it a suitable candidate for diabetic retinopathy detection. By combining it with the IFHO, the feature selection process has been optimized, which is essential for identifying relevant patterns and abnormalities related to diabetic retinopathy. The Diabetic Retinopathy 2015 dataset has been used to evaluate the effectiveness of the MobileNet-V2/IFHO model. The study results indicate that the DRMNV2/IFHO model consistently outperforms other methods in terms of precision, accuracy, and recall. Specifically, the model achieves an average precision of 97.521 %, accuracy of 96.986 %, and recall of 98.543 %. Moreover, when compared to advanced techniques, the DRMNV2/IFHO model demonstrates superior performance in specificity, F1-score, and AUC, with average values of 97.233 %, 93.8 %, and 0.927, respectively. These results underscore the potential of the DRMNV2/IFHO model as a valuable tool for improving the accuracy and efficiency of DR diagnosis. Nevertheless, additional validation and testing on larger datasets are required to verify the model's effectiveness and robustness in real-world clinical scenarios.

YOLOv5s fabric defect detection model with mixed attention mechanism

In the field of industrial quality inspection, fabric defect detection is regarded as an important research topic. The traditional inspection method mainly relies on manual visual inspection, which is time-consuming and easy to be affected by human factors, and has the problems of low accuracy, slow speed and high cost. Therefore, it is very important to use digital image processing technology and target detection technology to develop high precision, high efficiency and low-cost fabric defect detection model suitable for application deployment. Driven by deep learning-based neural network technology, many models with excellent performance have emerged in the field of object detection. This paper will improve on the basis of YOLOv5s model. ACmix module is introduced in Backbone layer of YOLOv5s model. This module overcomes the problem that convolution operation may cause local information loss after feature extraction, improves the model's limitation on feature extraction of complex background images, and enhances the model's processing ability in local feature fusion. Compared with the original YOLOv5s algorithm, mAP has improved by 0.6%. In the complex background, the detection accuracy of fabric defects targets has been significantly optimized and improved.

ScNovel: a scalable deep learning-based network for novel rare cell discovery in single-cell transcriptomics.

Single-cell RNA sequencing has achieved massive success in biological research fields. Discovering novel cell types from single-cell transcriptomics has been demonstrated to be essential in the field of biomedicine, yet is time-consuming and needs prior knowledge. With the unprecedented boom in cell atlases, auto-annotation tools have become more prevalent due to their speed, accuracy and user-friendly features. However, existing tools have mostly focused on general cell-type annotation and have not adequately addressed the challenge of discovering novel rare cell types. In this work, we introduce scNovel, a powerful deep learning-based neural network that specifically focuses on novel rare cell discovery. By testing our model on diverse datasets with different scales, protocols and degrees of imbalance, we demonstrate that scNovel significantly outperforms previous state-of-the-art novel cell detection models, reaching the most AUROC performance(the only one method whose averaged AUROC results are above 94%, up to 16.26% more comparing to the second-best method). We validate scNovel's performance on a million-scale dataset to illustrate the scalability of scNovel further. Applying scNovel on a clinical COVID-19 dataset, three potential novel subtypes of Macrophages are identified, where the COVID-related differential genes are also detected to have consistent expression patterns through deeper analysis. We believe that our proposed pipeline will be an important tool for high-throughput clinical data in a wide range of applications.

Projections of future spatiotemporal urban 3D expansion in China under shared socioeconomic pathways

The 21st century is marked by urbanization, with global focus on horizontal city expansion. Yet, vertical growth lacks research due to data scarcity. This study investigates the fusion of urban building height predictions with conventional sprawl projections to offer a holistic 3D urban expansion foresight. We constructed a model for projecting urban building volume using socio-economic factors includes GDP, population, and urbanization indicators, enabling the anticipation of China's 2010–2100 building volume demand across five Shared Socioeconomic Pathways (SSPs). Furthermore, deep learning-based neural networks were harnessed for estimating land conversion probabilities and building heights. By aggregating the projected demand, we estimated spatial expansion and building heights for each scenario at ten-year intervals from 2010 to 2100. Compared to other products, our proposed method excels in spatial granularity and temporal continuity, which includes height properties in the 3D structure. According to the projections of this study, considering various SSPs, China's building volume is expected to reach approximately 4500 km3 to 6500 km3 by 2050, representing 1.3–1.9 times increase compared to 2010. Furthermore, forecasts for 2050–2100 anticipate a notable decline to 60 % of the 2050 demand. At a more local level, potential spatial imbalances may arise, as the projections indicate significant urban expansion in the eastern areas and a heightened density of urban land within existing city clusters. This approach surpasses the constraints of 2D analysis, pioneering the prediction of fine-resolution, long-term spatiotemporal urban expansion in China (2010–2100), contributing a model and data support for future sustainable urban development pursuits.

Deep Learning–Augmented ECG Analysis for Screening and Genotype Prediction of Congenital Long QT Syndrome

Congenital long QT syndrome (LQTS) is associated with syncope, ventricular arrhythmias, and sudden death. Half of patients with LQTS have a normal or borderline-normal QT interval despite LQTS often being detected by QT prolongation on resting electrocardiography (ECG). To develop a deep learning-based neural network for identification of LQTS and differentiation of genotypes (LQTS1 and LQTS2) using 12-lead ECG. This diagnostic accuracy study used ECGs from patients with suspected inherited arrhythmia enrolled in the Hearts in Rhythm Organization Registry (HiRO) from August 2012 to December 2021. The internal dataset was derived at 2 sites and an external validation dataset at 4 sites within the HiRO Registry; an additional cross-sectional validation dataset was from the Montreal Heart Institute. The cohort with LQTS included probands and relatives with pathogenic or likely pathogenic variants in KCNQ1 or KCNH2 genes with normal or prolonged corrected QT (QTc) intervals. Convolutional neural network (CNN) discrimination between LQTS1, LQTS2, and negative genetic test results. The main outcomes were area under the curve (AUC), F1 scores, and sensitivity for detecting LQTS and differentiating genotypes using a CNN method compared with QTc-based detection. A total of 4521 ECGs from 990 patients (mean [SD] age, 42 [18] years; 589 [59.5%] female) were analyzed. External validation within the national registry (101 patients) demonstrated the CNN's high diagnostic capacity for LQTS detection (AUC, 0.93; 95% CI, 0.89-0.96) and genotype differentiation (AUC, 0.91; 95% CI, 0.86-0.96). This surpassed expert-measured QTc intervals in detecting LQTS (F1 score, 0.84 [95% CI, 0.78-0.90] vs 0.22 [95% CI, 0.13-0.31]; sensitivity, 0.90 [95% CI, 0.86-0.94] vs 0.36 [95% CI, 0.23-0.47]), including in patients with normal or borderline QTc intervals (F1 score, 0.70 [95% CI, 0.40-1.00]; sensitivity, 0.78 [95% CI, 0.53-0.95]). In further validation in a cross-sectional cohort (406 patients) of high-risk patients and genotype-negative controls, the CNN detected LQTS with an AUC of 0.81 (95% CI, 0.80-0.85), which was better than QTc interval-based detection (AUC, 0.74; 95% CI, 0.69-0.78). The deep learning model improved detection of congenital LQTS from resting ECGs and allowed for differentiation between the 2 most common genetic subtypes. Broader validation over an unselected general population may support application of this model to patients with suspected LQTS.

An ensemble-based approach for short-term load forecasting for buildings with high proportion of renewable energy sources

The increasing integration of renewable energy sources into large buildings and structures has emphasized the importance of effective control systems to optimize resource use, grid stability, and reliability. In this context, load forecasting is critical for predicting future energy demand and managing the intermittency and variability of renewable energy resources to ensure a stable performance. This study proposes an ensemble-based short-term load forecasting (STLF) method tailored for buildings. It combines parallel and series approaches for a 24-hour forecasting horizon. The parallel approach uses similar historical days to capture the general load trend, holidays, and user behavior, while the series component employs a deep learning-based neural network to address short-term changes. The final ensemble-based forecast is the culmination of these two methods. Evaluated using real load profile data sets with 15-minute resolution, the proposed method demonstrates more than 23% and 24% improvements in MAE and RMSE, respectively, compared to other methods. These results demonstrated the superior performance of the proposed method in terms of both accuracy and robustness, making it an effective solution for short-term load forecasting in buildings heavily reliant on renewable energy.

Multi-modal brain tumor segmentation via conditional synthesis with Fourier domain adaptation

Accurate brain tumor segmentation is critical for diagnosis and treatment planning, whereby multi-modal magnetic resonance imaging (MRI) is typically used for analysis. However, obtaining all required sequences and expertly labeled data for training is challenging and can result in decreased quality of segmentation models developed through automated algorithms.In this work, we examine the possibility of employing a conditional generative adversarial network (GAN) approach for synthesizing multi-modal images to train deep learning-based neural networks aimed at high-grade glioma (HGG) segmentation. The proposed GAN is conditioned on auxiliary brain tissue and tumor segmentation masks, allowing us to attain better accuracy and control of tissue appearance during synthesis. To reduce the domain shift between synthetic and real MR images, we additionally adapt the low-frequency Fourier space components of synthetic data, reflecting the style of the image, to those of real data. We demonstrate the impact of Fourier domain adaptation (FDA) on the training of 3D segmentation networks and attain significant improvements in both the segmentation performance and prediction confidence. Similar outcomes are seen when such data is used as a training augmentation alongside the available real images. In fact, experiments on the BraTS2020 dataset reveal that models trained solely with synthetic data exhibit an improvement of up to 4% in Dice score when using FDA, while training with both real and FDA-processed synthetic data through augmentation results in an improvement of up to 5% in Dice compared to using real data alone. This study highlights the importance of considering image frequency in generative approaches for medical image synthesis and offers a promising approach to address data scarcity in medical imaging segmentation.

Effect of Different Modalities of Facial Images on ASD Diagnosis using Deep Learning-Based Neural Network

This paper aims to investigate the effectiveness of different modalities of facial images for diagnosing Autism Spectrum Disorder (ASD) using deep learning-based neural networks. The motivation behind this study is the potential of advanced technologies to aid in accurately diagnosing ASD. The research revolves around the need to explore the performance of deep learning models on different modalities of facial images and to identify the challenges and potential solutions associated with each modality. The methodology involves training and testing the models on the respective datasets and analysing their accuracy and performance. ResNet50V2 achieved a 100% accuracy on the 2D test dataset, while Xception achieved an accuracy of 93.75% on the 3D test set. The detection accuracy suggests that neural networks-based deep learning methods have the potential to diagnose ASD using facial images accurately. However, the models perform better on 2D data, highlighting the need for additional training on larger 3D datasets to improve accuracy on 3D images. The study contributes to the field by providing insights into the performance of different modalities of facial images, emphasizing the need for robust datasets, and suggesting future research directions to enhance the accuracy and efficiency of ASD diagnosis using deep learning techniques.

Robust neural decoding for dexterous control of robotic hand kinematics

BackgroundManual dexterity is a fundamental motor skill that allows us to perform complex daily tasks. Neuromuscular injuries, however, can lead to the loss of hand dexterity. Although numerous advanced assistive robotic hands have been developed, we still lack dexterous and continuous control of multiple degrees of freedom in real-time. In this study, we developed an efficient and robust neural decoding approach that can continuously decode intended finger dynamic movements for real-time control of a prosthetic hand. MethodsHigh-density electromyogram (HD-EMG) signals were obtained from the extrinsic finger flexor and extensor muscles, while participants performed either single-finger or multi-finger flexion-extension movements. We implemented a deep learning-based neural network approach to learn the mapping from HD-EMG features to finger-specific population motoneuron firing frequency (i.e., neural-drive signals). The neural-drive signals reflected motor commands specific to individual fingers. The predicted neural-drive signals were then used to continuously control the fingers (index, middle, and ring) of a prosthetic hand in real-time. ResultsOur developed neural-drive decoder could consistently and accurately predict joint angles with significantly lower prediction errors across single-finger and multi-finger tasks, compared with a deep learning model directly trained on finger force signals and the conventional EMG-amplitude estimate. The decoder performance was stable over time and was robust to variations of the EMG signals. The decoder also demonstrated a substantially better finger separation with minimal predicted error of joint angle in the unintended fingers. ConclusionsThis neural decoding technique offers a novel and efficient neural-machine interface that can consistently predict robotic finger kinematics with high accuracy, which can enable dexterous control of assistive robotic hands.

Neural network-based model for evaluating inert nodules and volume doubling time in T1 lung adenocarcinoma: a nested case-control study.

The purpose of this study is to establish model for assessing inert nodules predicting nodule volume-doubling. A total of 201 patients with T1 lung adenocarcinoma were analysed retrospectively pulmonary nodule information was predicted by an AI pulmonary nodule auxiliary diagnosis system. The nodules were classified into two groups: inert nodules (volume-doubling time (VDT)>600 days n=152) noninert nodules (VDT<600 days n=49). Then taking the clinical imaging features obtained at the first examination as predictive variables the inert nodule judgement model <sn</sn>>(INM) volume-doubling time estimation model (VDTM) were constructed based on a deep learning-based neural network. The performance of the INM was evaluated by the area under the curve (AUC) obtained from receiver operating characteristic (ROC) analysis the performance of the VDTM was evaluated by R2(determination coefficient). The accuracy of the INM in the training and testing cohorts was 81.13% and 77.50%, respectively. The AUC of the INM in the training and testing cohorts was 0.7707 (95% CI 0.6779-0.8636) and 0.7700 (95% CI 0.5988-0.9412), respectively. The INM was effective in identifying inert pulmonary nodules; additionally, the R2 of the VDTM in the training cohort was 0.8008, and that in the testing cohort was 0.6268. The VDTM showed moderate performance in estimating the VDT, which can provide some reference during a patients' first examination and consultation. The INM and the VDTM based on deep learning can help radiologists and clinicians distinguish among inert nodules and predict the nodule volume-doubling time to accurately treat patients with pulmonary nodules.

Adoption of a deep learning-based neural network model in the psychological behavior analysis of resident tourism consumption.

With the development of society and the continuous progress of science and technology, it has become the mainstream measure to promote the development of the social economy through science and technology. Therefore, to improve the current situation of tourism consumption, improve the consumer sentiment of tourists, and promote the development of the tourism economy, the convolutional neural network (CNN) technology model is used to analyze the tourist's consumer psychology and behavior. Based on this, the user's consumption situation is analyzed, thus providing support for the intelligent improvement of tourism consumption. First, the basic characteristics of tourism consumption mood and behavior are introduced, and the methods to improve the tourism consumption mood and behavior are briefly introduced. Then, the CNN algorithm is employed to identify consumers' travel consumption behaviors and emotions. To improve the recognition effect, the algorithm is combined with skeleton node behavior recognition and video image behavior recognition. Finally, the performance of the designed algorithm is tested. The accuracy of the human behavior recognition (HBR) algorithm is more than 0.88. Compared with the detection effect of the HBR algorithm, the combined algorithm adopted in this work can reduce the image processing time and improve the detection efficiency. The multithread method can effectively reduce the complexity of the model and improve the recognition accuracy. The test results on different data sets show that the proposed algorithm can better adapt to the changes in identification samples and obtain more accurate recognition results compared with similar algorithms. In summary, this study not only provides technical support for the rational analysis of consumer sentiment and consumer behavior but also contributes to the comprehensive development of the tourism market.

BUS-Net: Breast Tumour Detection Network for Ultrasound Images Using Bi-directional ConvLSTM and Dense Residual Connections.

Breast ultrasound (BUS) imaging has become one of the key imaging modalities for medical image diagnosis and prognosis. However, the manual process of lesion delineation from ultrasound images can incur various challenges concerning variable shape, size, intensity, curvature, or other medical priors of the lesion in the image. Therefore, computer-aided diagnostic (CADx) techniques incorporating deep learning-based neural networks are automatically used to segment the lesion from BUS images. This paper proposes an encoder-decoder-based architecture to recognize and accurately segment the lesion from two-dimensional BUS images. The architecture is utilized with the residual connection in both encoder and decoder paths; bi-directional ConvLSTM (BConvLSTM) units in the decoder extract the minute and detailed region of interest (ROI) information. BConvLSTM units and residual blocks help the network weigh ROI information more than the similar background region. Two public BUS image datasets, one with 163 images and the other with 42 images, are used. The proposed model is trained with the augmented images (ten forms) of dataset one (with 163 images), and test results are produced on the second dataset and the testing set of the first dataset-the segmentation performance yielding comparable results with the state-of-the-art segmentation methodologies. Similarly, the visual results show that the proposed approach for BUS image segmentation can accurately identify lesion contours and can potentially be applied for similar and larger datasets.

AI-CardioCare: Artificial Intelligence Based Device for Cardiac Health Monitoring

Cardiac disorders are one of the leading causes of mortality around the globe and early diagnosis of heart diseases can be beneficial for its mitigation. In this article, an artificial intelligence (AI) based device has been proposed, which allows for an automatic and real-time diagnosis of cardiac diseases based on deep learning techniques. The heart sound (phonocardiogram) signal is acquired by a customized designed stethoscope and the signal is processed before analysis using AI methods for the classification of four major cardiac diseases (Aortic Stenosis, Mitral Regurgitation, Mitral Stenosis, and Mitral Valve Prolapse). Two deep learning-based neural networks, one-dimensional (1-D) convolutional neural network (CNN) and spectrogram based 2-D-CNN models from the analysis of these signals has been integrated with a low-cost single-board processor to make a standalone device. All data processing is done in a single hardware setup and user interface is provided allowing the user to control the data accessibility and visibility to generate the diagnostic report. As a result, the developed device has demonstrated to be a valuable low-cost diagnostic tool for both medical professionals and personal usage at home.

Evaluating the difference in walk patterns among normal-weight and overweight/obese individuals in real-world surfaces using statistical analysis and deep learning methods with inertial measurement unit data.

Unusual walk patterns may increase individuals' risks of falling. Anthropometric features of the human body, such as the body mass index (BMI), influences the walk patterns of individuals. In addition to the BMI, uneven walking surfaces may cause variations in the usual walk patterns of an individual that will potentially increase the individual's risk of falling. The objective of this study was to statistically evaluate the variations in the walk patterns of individuals belonging to two BMI groups across a wide range of walking surfaces and to investigate whether a deep learning method could classify the BMI-specific walk patterns with similar variations. Data collected by wearable inertial measurement unit (IMU) sensors attached to individuals with two different BMI were collected while walking on real-world surfaces. In addition to traditional statistical analysis tools, an advanced deep learning-based neural network was used to evaluate and classify the BMI-specific walk patterns. The walk patterns of overweight/obese individuals showed a greater correlation with the corresponding walking surfaces than the normal-weight population. The results were supported by the deep learning method, which was able to classify the walk patterns of overweight/obese (94.8 ± 4.5%) individuals more accurately than those of normal-weight (59.4 ± 23.7%) individuals. The results suggest that application of the deep learning method is more suitable for recognizing the walk patterns of overweight/obese population than those of normal-weight individuals. The findings from the study will potentially inform healthcare applications, including artificial intelligence-based fall assessment systems for minimizing the risk of fall-related incidents among overweight and obese individuals.

Deep learning-based neural networks for day-ahead power load probability density forecasting.

Energy efficiency is crucial to greenhouse gas (GHG) emission pathways reported by the Intergovernmental Panel on Climate Change. Electrical overload frequently occurs and causes unwanted outages in distribution networks, which reduces energy utilization efficiency and raises environmental risks endangering public safety. Electrical load, however, has a dynamically fluctuating behavior with notoriously nonlinear hourly, daily, and seasonal patterns. Accurate and reliable load forecasting plays an important role in scheduling power generation processes and preventing electrical systems from overloading; nevertheless, such forecasting is fundamentally challenging, especially under highly variable power load and climate conditions. This study proposed a deep learning-based monotone composite quantile regression neural network (D-MCQRNN) model to extract the multiple non-crossing and nonlinear quantile functions while conquering the drawbacks of error propagation and accumulation encountered in multi-step-ahead probability density forecasting. The constructed models were assessed by an hourly power load series collected at the electric grid center of Henan Province in China in two recent years, along with the corresponding meteorological data collected at 16 monitoring stations. The results demonstrated that the proposed D-MCQRNN model could significantly alleviate the time-lag and biased-prediction phenomena and noticeably improve the accuracy and reliability of multi-step-ahead probability density forecasts on power load. Consequently, the proposed model can significantly reduce the risk and impact of overload faults and effectively promote energy utilization efficiency, thereby mitigating GHG emissions and moving toward cleaner energy production.

CAR-Net: A Deep Learning-Based Deformation Model for 3D/2D Coronary Artery Registration.

Percutaneous coronary intervention is widely applied for the treatment of coronary artery disease under the guidance of X-ray coronary angiography (XCA) image. However, the projective nature of XCA causes the loss of 3D structural information, which hinders the intervention. This issue can be addressed by the deformable 3D/2D coronary artery registration technique, which fuses the pre-operative computed tomography angiography volume with the intra-operative XCA image. In this study, we propose a deep learning-based neural network for this task. The registration is conducted in a segment-by-segment manner. For each vessel segment pair, the centerlines that preserve topological information are decomposed into an origin tensor and a spherical coordinate shape tensor as network input through independent branches. Features of different modalities are fused and processed for predicting angular deflections, which is a special type of deformation field implying motion and length preservation constraints for vessel segments. The proposed method achieves an average error of 1.13 mm on the clinical dataset, which shows the potential to be applied in clinical practice.

Effects of Temperature Rise on Clean Energy-Based Capital Market Investments: Neural Network-Based Granger Causality Analysis

During the past 20 years, due to climate change, the government and the private sector have significantly focused on relying on non-fossil fuel-based methods for their energy needs. Climate change-related events, such as unusual weather conditions, abnormal temperature spikes, etc., have an adverse influence on clean energy-based investments. In the given study, we intend to focus on how an incremental temperature rise could affect investors’ perceptions of clean energy assets. To understand the investor-based sentiment on climate change, we utilize prominent clean energy ETFs (exchange traded funds) and consider the temperature’s effect on them. The daily average temperatures of the three most dynamic international financial centers: New York, London and Tokyo, are taken as predictors. Deep learning-based neural networks are applied to understand both the linear and non-linear relationships between the desired variables and identify the causal effects. The results indicate that in almost all the cases with desired lags, there is some sort of non-linear causality, irrespective of linear causality effects. We hope this occurrence can help portfolio managers and environmental professionals in identifying novel climate change-related factors when considering the temperature-related risks.

Time series-based groundwater level forecasting using gated recurrent unit deep neural networks

ABSTRACT In this research, the mean monthly groundwater level with a range of 3.78 m in Qoşaçay plain, Iran, is forecast. Regarding three different layers of gated recurrent unit (GRU) structures and a hybrid of variational mode decomposition with gated recurrent unit (VMD-GRU), deep learning-based neural network models are developed. As the base model for performance comparison, the general single-long short-term memory-layer network model is developed. In all models, the module of sequence-to-one is used because of the lack of meteorological variables recorded in the study area. For modeling, 216 monthly datasets of the mean monthly water table depth of 33 different monitoring piezometers in the period April 2002–March 2020 are utilized. To boost the performance of the models and reduce the overfitting problem, an algorithm tuning process using different types of hyperparameter accompanied by a trial-and-error procedure is applied. Based on performance evaluation metrics, the total learnable parameters value and especially the model grading process, the new double-GRU model coupled with multiplication layer (×) (GRU2× model) is chosen as the best model. Under the optimal hyperparameters, the GRU2× model results in an R 2 of 0.86, a root mean square error (RMSE) of 0.18 m, a corrected Akaike’s information criterion (AICc) of −280.75, a running time for model training of 87 s and a total grade (TG) of 6.21 in the validation stage; and the hybrid VMD-GRU model yields an RMSE of 0.16 m, an R 2 of 0.92, an AICc of −310.52, a running time of 185 s and a TG of 3.34.