Recurrent Neural Network Algorithm Research Articles

Heteroscedasticity effects as component to future stock market predictions using RNN-based models.

Heteroscedasticity effects are useful for forecasting future stock return volatility. Stock volatility forecasting provides business insight into the stock market, making it valuable information for investors and traders. Predicting stock volatility is a crucial task and challenging. This study proposes a hybrid model that predicts future stock volatility values by considering the heteroscedasticity element of the stock price. The proposed model is a combination of Generalized Autoregressive Conditional Heteroskedasticity (GARCH) and a well-known Recurrent Neural Network (RNN) algorithm Long Short-Term Memory (LSTM). This proposed model is referred to as GARCH-LSTM model. The proposed model is expected to improve prediction accuracy by considering heteroscedasticity elements. First, the GARCH model is employed to estimate the model parameters. After that, the ARCH effect test is used to test the residuals obtained from the model. Any untrained heteroscedasticity element must be found using this step. The hypothesis of the ARCH test yielded a p-value less than 0.05 indicating there is valuable information remaining in the residual, known as heteroscedasticity element. Next, the dataset with heteroscedasticity is then modelled using an LSTM-based RNN algorithm. Experimental results revealed that hybrid GARCH-LSTM had the lowest MAE (7.961), RMSE (10.466), MAPE (0.516) and HMAE (0.005) values compared with a single LSTM. The accuracy of forecasting was also significantly improved by 15% and 13% with hybrid GARCH-LSTM in comparison to single LSTMs. Furthermore, the results reveal that hybrid GARCH-LSTM fully exploits the heteroscedasticity element, which is not captured by the GARCH model estimation, outperforming GARCH models on their own. This finding from this study confirmed that hybrid GARCH-LSTM models are effective forecasting tools for predicting stock price movements. In addition, the proposed model can assist investors in making informed decisions regarding stock prices since it is capable of closely predicting and imitating the observed pattern and trend of KLSE stock prices.

Enhancing Durrotalk Chatbot Accuracy Utilizing a Hybrid Model Based on Recurrent Neural Network (RNN) Algorithm and Decision Tree

DurroTalk, a chatbot for new student admissions at Pondok Pesantren Durrotu Ahlissunnah Waljamaah, Semarang, integrates a hybrid model with Recurrent Neural Network (RNN) and Decision Tree. RNN, the base model, employs Natural Language Processing (NLP) to understand sentence structure and context, overcoming vanishing gradient through LSTM layers. The Decision Tree normalizes words, addressing slang and synonyms. The hybrid model boosts chatbot accuracy by 9%, reaching 77% from the initial 68%. This research signifies progress in integrating artificial intelligence into traditional education, showcasing a chatbot adept at handling non-standard language. Decision Tree integration enhances overall performance, making the chatbot proficient in understanding user inputs and generating contextually relevant responses. This study exemplifies the potential of AI, particularly chatbot technology, in modernizing educational processes at traditional institutions.

Music Teaching Mode of Colleges and Universities Based On Hierarchically Gated Recurrent Neural Network (HGRNN) and Lyrebird Optimization Algorithm (LOA)

Colleges and universities play a crucial role in nurturing talent and providing highly skilled individuals for various sectors of society. Through modifications over time, the model of music education at colleges and universities has advanced. However, there are still numerous issues that demand careful consideration. This manuscript proposes a hierarchically gated recurrent neural network (HGRNN) optimized with the lyrebird Optimization Algorithm (LOA) for predicting music teaching mode of colleges and universities (MTM-HGRNN-LOA). Initially, the data is collected via real time basis. Afterward, the data is fed to an unscented trainable kalman filter (UTKF) based pre-processing process. In the pre-processing segment, it enhances training rate and eliminates the of batch size dependency. The pre-processing output is given to modified spline-kernelled chirp let transform (MSKCT). The input signal undergoes feature extraction to derive the primary features, which are subsequently combined to yield more comprehensive features in an efficient manner. After that, the extracted features are given to a hierarchically gated recurrent neural network and lyrebird optimization algorithm for effectively classifying the music teaching mode for best, good, normal, satisfactory and poor. The weight parameters of hierarchically gated recurrent neural network are optimized using the lyrebird optimization algorithm. The proposed method is implemented in python and evaluated their performance with existing methods. The performance metrics, like precision, F1-score, accuracy, specificity, sensitivity, and ROC is analysed to the proposed method's performance. The proposed MTM-HGRNN-LOA methods of accuracy are provide 97% best, 98% good, 95% normal, 98% satisfactory and 97% poor music teaching mode. The existing methods MTM-CNN, MTM-BPNN and MTM-GNN, the specificity becomes 90%, 70%, 79% best, 77%, 75%, 65% good, 66%, 85%, 84% normal, 59%, 58%, 70% satisfactory, 61%, 79%, 81% poor music teaching mode. The results show that the proposed MTM-HGRNN-LOA method outperforms other existing techniques, such as online vocal music teaching quality using Back Propagation neural network and convolutional neural network based College-Level Music Teaching Quality Evaluation.

LSTM RNN-based excitation force prediction for the real-time control of wave energy converters

Wave energy is a type of abundant and dense renewable energy. Wave force prediction is a critical technology that influences power absorption efficiency in the real-time control of wave energy converter (WEC). Could wave elevation be used to predict wave excitation force directly by training artificial neural network? This method results in rapid and suitable prediction for real-time control. A long short-term memory recurrent neural network (LSTM RNN) algorithm is introduced to identify characteristics of wave excitation forces based on wave elevations. In this method, the wave elevations in front of the structure are measured to obtain sufficient time to actuate the control manipulation. A total of 180 regular wave and 12 irregular wave tests are conducted, and the LSTM RNN model is trained based on the experimental results. The performance of the LSTM algorithm is verified. According to the regular cases in the study, the LSTM prediction can identify high-order harmonic loads, and the anti-noise capability of the LSTM algorithm can filter random noises from the measure signals. In the irregular cases, the LSTM RNN algorithm performs effectively to predict the wave force excited on the structure using wave elevations measured by wave probes. The best combinations of the test setting parameters are determined to guide experimental tests and WEC prototypes.

An intelligent error correction model for English grammar with hybrid attention mechanism and RNN algorithm

Abstract This article proposes an English grammar intelligent error correction model based on the attention mechanism and Recurrent Neural Network (RNN) algorithm. It aims to improve the accuracy and effectiveness of error correction by combining the powerful context-capturing ability of the attention mechanism with the sequential modeling ability of RNN. First, based on the improvement of recurrent neural networks, a bidirectional gated recurrent network is added to form a dual encoder structure. The encoder is responsible for reading and understanding the input text, while the decoder is responsible for generating the corrected text. Second, the attention mechanism is introduced into the decoder to convert the output of the encoder into the attention probability distribution for integration. This allows the model to focus on the relevant input word as it generates each corrected word. The results of the study showed that the model was 2.35% points higher than statistical machine translation–neural machine translation in the CoNLL-2014 test set, and only 1.24 points lower than the human assessment score, almost close to the human assessment level. The model proposed in this study not only created a new way of English grammar error correction based on the attention mechanism and RNN algorithm in theory but also effectively improved the accuracy and efficiency of English grammar error correction in practice. It further provides English learners with higher-quality intelligent error correction tools, which can help them learn and improve their English level more effectively.

DWT-BILSTM-based models for day-ahead hourly global horizontal solar irradiance forecasting

AbstractAccurate forecasting of electricity generation from renewable energy sources is crucial for the operation, planning and management of smart grids. For reliable planning and operation of photovoltaic (PV) systems in grid-connected or islanded utilities, an hourly day-ahead forecast of PV output is critical. The forecast of PV power can be done indirectly by estimating solar irradiance. For forecasting day-ahead hourly global horizontal irradiance (GHI), two forecasting models with different multivariate inputs are proposed in this paper, and the results are compared. These models use a hybrid algorithm of discrete wavelet decomposition and bidirectional long short-term memory (BILSTM). The inputs of the first model contain GHI and weather type data. The other model allows for observation of the effect of meteorological values including GHI, temperature, humidity, wind speed, and weather type data. The forecasting performance of deep learning algorithms which contain recurrent neural network (RNN), long short-term memory (LSTM), and BILSTM algorithms for day ahead hourly solar irradiance forecasting problems are also compared. To evaluate the performance of proposed models, two datasets are used for Model 1 and one dataset is used for Model 2. An experiment is also done to demonstrate that the proposed Model 1 is applicable in datasets collected in the vicinity of the city of Trabzon. On the other hand, BILSTM algorithm outperforms RNN and LSTM algorithms. It is seen that the test successes of both proposed models are better than the results given in the literature.

Fuzzy Deep Learning Recurrent Neural Network Algorithm to Detect Corn Leaf Disease

Corn is a major commodity after rice in supporting food self-sufficiency in Indonesia. However, due to leaf disease, the quality and quantity of corn plants are greatly reduced. The problem with detecting corn leaf diseases is that the detection method is still manual, making it inefficient and ineffective. Therefore, in this study, disease detection on corn leaves was performed using the Fuzzy C-Means (FCM) and Long Short-Term Memory (LSTM) methods. First, oversampling was carried out to ensure an equal amount of data in all classes, then the corn leaf images were pre-processed before being input into the LSTM algorithm. After completing clustering process in the [Formula: see text] algorithm, the next step involved extracting texture features using the Gray Level Co-occurrence Matrix (GLCM) technique, followed by classification using LSTM. To assess their performance, both algorithms underwent evaluation using the k-fold cross-validation method, and their accuracy and speed were compared. The results of the k-fold cross-validation demonstrated that the [Formula: see text] algorithm achieved an accuracy of 63.53%, whereas the LSTM algorithm achieved an accuracy of 80.24%. In terms of the time required for training and prediction, the LSTM algorithm took 13[Formula: see text]min and 18[Formula: see text]s for training on corn leaf disease images, while the prediction process only took 1.59[Formula: see text]s. The training and prediction time required for the [Formula: see text] algorithm were 65[Formula: see text]min and 24[Formula: see text]s and 5[Formula: see text]min and 44[Formula: see text]s, respectively. The conclusion of this study is that the LSTM algorithm has better accuracy and time compared to [Formula: see text] on the dataset used in this study in terms of corn leaf disease detection.

An improved analogy-rule based software effort estimation using HTRR-RNN in software project management

To complete a software project, the most realistic effort required is predicted by the process called Software Effort Estimation (SEE). Effort Estimation (EE) in software development is considered a complicated task owing to the variability of variables. Thus, numerous Software Development EE (SDEE) models have been developed; nevertheless, these models' accuracy rates are not good enough, and the variations among software functionalities were not focused for SEE. Therefore, by considering it as the research gap in Software Project Management (SPM), an enhanced analogy-rule-centric SEE utilizing Hyperbolic Tangent Recurrent Radial − Recurrent Neural Networks (HTRR-RNN) algorithm is proposed in this work. Initially, regarding the project size, the projects are partitioned. Next, the highly vital features are extracted as of the separated projects. By employing Levy Flight-Rock Hyraxes Swarm (LH-RHS) optimization, the selective quantity of features is chosen from the extracted features. By deploying the Mamdani Method − Fuzzy (MM-Fuzzy) algorithm, the rules are created regarding the selected features. Lastly, the number of efforts needed for the software development process is determined by the classification algorithm regarding the rules being created. COCOMO81, MAXWELL, and China are the openly accessible datasets utilized in this methodology. Then, the proposed method’s results are contrasted with the prevailing algorithms. As per the experimental results, the proposed methodology performs SEE with a better accuracy rate; thus, it surpassed the other existing methodologies.

Design of an English vocabulary e-learning recommendation system based on word bag model and recurrent neural network algorithm

This article takes learning English vocabulary as the foundation, constructs precise classification strategies, and designs a recommendation system for English vocabulary learning. This article constructs a recommendation system for English vocabulary learning based on the word bag calculation model and recursive neural network calculation method, which has great significance, and also conducts in-depth research on this system. Thus, a classification calculation method based on the word bag calculation model was proposed. The calculation method in this paper is different from traditional calculation methods, and the classification accuracy in this paper is higher. Meanwhile, recurrent neural networks themselves have one or more feedback loops that can feed back information from the neural network to other neural networks, resulting in recurrent networks with different structures. After a series of research results, it has been found that the SIFT method takes longer to extract than the SURF method, and the number of extracted features is relatively large. Therefore, the recommendation system for English vocabulary learning proposed in this article can be improved according to the different needs of users, in order to better meet their individual requirements.

An improved electric equipment fault prediction based on recursive multi-parameter prediction algorithm

This paper proposes a method of beforehand prediction of electric equipment faults based on chain-linked recurrent neural network algorithm, which takes the operating parameters of power equipment and other relevant environmental factors as inputs, and takes the fault characteristics as output judgment marks, and constructs a machine learning training model to realize the prediction of power equipment faults. The neural network algorithm adopted in this paper adopts a tree structure. Each sub-node can transfer information with its multiple superior nodes, so that the correlation between the data of the front and back nodes can be obtained, which meets the needs of the equipment fault prediction model. Considering that the occurrence of power transformer faults is sudden and greatly affected by changes in the surrounding environment, the input of prediction algorithms should consider more environmental factors. This method takes the historical data of various parameters including meteorological phenomena, geography data, and temperature of adjacent equipment and facilities as the training sample set, improves the learning model, gives the trend curve of each index, and gives a prompt at its threshold to ensure the prediction accuracy and give the index prediction.

State-of-charge estimation of lithium-ion battery: Joint long short-term memory network and adaptive extended Kalman filter online estimation algorithm

With the widespread use of lithium-ion batteries, estimating the State of Charge (SOC) has been one of the most critical tasks in the battery management system (BMS). In this paper, a joint long short-term memory recurrent neural network (LSTM-RNN) and adaptive extended Kalman filter (AEKF) algorithm is proposed to achieve accurate SOC estimation. The proposed algorithm generates an initial SOC estimation value through a trained LSTM-RNN with a sliding time window signal sequence as input. The initial estimation value is then used as a feedback input to the AEKF. The results show that only a 30-s time series as the input of the LSTM-RNN can achieve satisfactory estimation results. Compared with traditional adaptive extended Kalman filter and unscented Kalman filter (UKF), the proposed algorithm has a better performance of estimation accuracy under various discharging conditions, especially when applied at low temperatures. Finally, the SOC estimation results under four operating conditions and various temperatures show that the maximum error is less than 2.5 %, in particular, the root mean square error (RMSE) and the mean absolute error (MAE) are below 1.01 % and 0.64 %. The proposed algorithm is an effective SOC estimation method with good robustness and high estimation accuracy.

Improving respiratory signal prediction with a deep neural network and simple changes to the input and output data format

Objective. To improve respiratory gating accuracy and radiation treatment throughput, we developed a generalized model based on a deep neural network (DNN) for predicting any given patient’s respiratory motion. Approach. Our model uses long short-term memory (LSTM) based on a recurrent neural network (RNN), and improves upon common techniques. The first improvement is that the data input is not a one-dimensional sequence, but two-dimensional block data. This shortens the input sequence length, reducing computation time. Second, the output is not a scalar, but a sequence prediction. This increases the amount of available data, allowing improved prediction accuracy. For training and evaluation of our model, 434 sets of real-time position management data were retrospectively collected from clinical studies. The data were separated in a ratio of 4:1, with the larger set used for training models and the remaining set used for testing. We measured the accuracy of respiratory signal prediction and amplitude-based gating with prediction windows equaling 133, 333, and 533 ms. This new model was compared with the original LSTM and a non-recurrent DNN model. Main results. The mean absolute errors with the prediction window at 133, 333 and 533 ms were 0.036, 0.084, 0.119 with our model; 0.049, 0.14, 0.246 with the original LSTM-based model; and 0.041, 0.119, 0.16 with the non-recurrent DNN model, respectively. The computation time were 0.66 ms with our model; 0.63 ms the original LSTM-based model; 1.60 ms the non-recurrent DNN model, respectively. The accuracies of amplitude-based gating with the same prediction window settings and a duty cycle of approximately 50% were 98.3%, 95.8% and 92.7% with our model, 97.6%, 93.9% and 87.2% with the original LSTM-based model; and 97.9%, 94.3% and 89.5% with the non-recurrent DNN model, respectively. Significance. Our RNN algorithm for respiratory signal prediction successfully estimated tumor positions. We believe it will be useful in respiratory signal prediction technology.

An Assessment of Geological Hazard Management using Dynamically Stabilized Recurrent Neural Network and Beluga Whale Optimization Algorithm

Mountainous region geological hazards are a leading cause of natural disasters, resulting in significant human and economic losses. Regional topography, landforms, lithology, plant life, geological circumstances, and meteorology all have a significant impact on their creation. Gansu, situated in the interior of Northwestern China, features Lanzhou as its capital and primary urban center, positioned in the southeast of the province. Geological risks, particularly landslides, mudslides, and avalanches, present significant challenges to Gansu Province. Consequently, local authorities are actively devising customized strategies to mitigate these hazards and foster sustainable development. In this research work, an Assessment of Geological Hazard Management using Dynamically Stabilized Recurrent Neural Network and Beluga Whale Optimization Algorithm (AGL-HM-DSRNN-BWOA) is proposed. Initially, the input raster data are gathered from the Normalised Difference Vegetation Index (NDVI) dataset. The input raster data is then pre-processed using Adaptive Actor-Critic Bilateral Filter (A2CBF) to reduce noises and increase the overall quality of the raster data. To classify the geological hazards, the pre-processed raster data are fed into a neural network named DSRNN. The geological hazard is accurately categorized as low risk, medium-low risk, medium-high risk, high risk using proposed DSRNN. In general, DSRNN does not express some adaption of optimization strategies for determining optimal parameters to promise exact classification for managing geological hazard by assessment. Therefore, Beluga Whale Optimization Algorithm (BWOA) is proposed to enhance weight parameter of DSRNN classifier, which precisely assess for managing the geological hazards. The efficiency of the proposed AGL-HM-DSRNN-BWOA approach is evaluated using a number of performance criteria, including accuracy, sensitivity, specificity, ROC, mean square error, root mean square error, mean absolute error. The proposed AGL-HM-DSRNN-BWOA method attains 22.36%, 25.42% and 18.17% higher accuracy, 21.26%, 15.42% and 19.27% higher sensitivity, 28.36%, 25.32% and 28.27% higher F-measure compared with existing methods, such as the Risk assessment and its influencing factors examination of geological hazards in typical mountain environment (RA-IFA-GH-TME), Feasibility study of land cover categorization under normalized difference vegetation index for landslide risk assessment(LCC-NDVI-LRA), and Multiple hazard exposure mapping under machine learning for Salzburg, Austria (MH-EM-SSA-ML) respectively.

Accelerated four-dimensional free-breathing whole-liver water-fat magnetic resonance imaging with deep dictionary learning and chemical shift modeling.

Multi-echo chemical-shift-encoded magnetic resonance imaging (MRI) has been widely used for fat quantification and fat suppression in clinical liver examinations. Clinical liver water-fat imaging typically requires breath-hold acquisitions, with the free-breathing acquisition method being more desirable for patient comfort. However, the acquisition for free-breathing imaging could take up to several minutes. The purpose of this study is to accelerate four-dimensional free-breathing whole-liver water-fat MRI by jointly using high-dimensional deep dictionary learning and model-guided (MG) reconstruction. A high-dimensional model-guided deep dictionary learning (HMDDL) algorithm is proposed for the acceleration. The HMDDL combines the powers of the high-dimensional dictionary learning neural network (hdDLNN) and the chemical shift model. The neural network utilizes the prior information of the dynamic multi-echo data in spatial respiratory motion, and echo dimensions to exploit the features of images. The chemical shift model is used to guide the reconstruction of field maps, maps, water images, and fat images. Data acquired from ten healthy subjects and ten subjects with clinically diagnosed nonalcoholic fatty liver disease (NAFLD) were selected for training. Data acquired from one healthy subject and two NAFLD subjects were selected for validation. Data acquired from five healthy subjects and five NAFLD subjects were selected for testing. A three-dimensional (3D) blipped golden-angle stack-of-stars multi-gradient-echo pulse sequence was designed to accelerate the data acquisition. The retrospectively undersampled data were used for training, and the prospectively undersampled data were used for testing. The performance of the HMDDL was evaluated in comparison with the compressed sensing-based water-fat separation (CS-WF) algorithm and a parallel non-Cartesian recurrent neural network (PNCRNN) algorithm. Four-dimensional water-fat images with ten motion states for whole-liver are demonstrated at several R values. In comparison with the CS-WF and PNCRNN, the HMDDL improved the mean peak signal-to-noise ratio (PSNR) of images by 9.93 and 2.20 dB, respectively, and improved the mean structure similarity (SSIM) of images by 0.058 and 0.009, respectively, at R=10. The paired t-test shows that there was no significant difference between HMDDL and ground truth for proton-density fat fraction (PDFF) and values at R up to 10. The proposed HMDDL enables features of water images and fat images from the highly undersampled multi-echo data along spatial, respiratory motion, and echo dimensions, to improve the performance of accelerated four-dimensional (4D) free-breathing water-fat imaging.

Unlocking the potential: CRNN-driven flexible mechanical sensors for intelligent sensing and beyond

In our rapidly evolving society, intelligent technology has become a driving force across various sectors, leading to transformative progress. This paper explores the integration of Convolutional Recurrent Neural Network (CRNN) algorithms with flexible sensors, showcasing their remarkable potential in enhancing data processing and interpretation. The synergy between CRNN and flexible sensors holds great promise in diverse domains, including health monitoring and industrial automation. Our comprehensive evaluation of the CRNN model reveals exceptional performance in classification and regression tasks, highlighting its adaptability to complex and varied sensor data. Additionally, confusion matrix analysis provides insights into multi-class classification scenarios, reinforcing the model's reliability. A comparative analysis against traditional machine learning techniques demonstrates the superiority of CRNN in handling time-series data. In conclusion, the integration of CRNN with flexible sensors is set to revolutionize intelligent sensing technology, opening new avenues for innovation and problem-solving in various industries. This collaborative approach not only enhances data accuracy and reliability but also paves the way for groundbreaking technologies and applications yet to be imagined. The future of intelligent sensing is here, and it is both exciting and promising, with CRNN-driven flexible sensors leading the way towards a more intelligent and connected world.

Exploiting Symmetric Temporally Sparse BPTT for Efficient RNN Training

Recurrent Neural Networks (RNNs) are useful in temporal sequence tasks. However, training RNNs involves dense matrix multiplications which require hardware that can support a large number of arithmetic operations and memory accesses. Implementing online training of RNNs on the edge calls for optimized algorithms for an efficient deployment on hardware. Inspired by the spiking neuron model, the Delta RNN exploits temporal sparsity during inference by skipping over the update of hidden states from those inactivated neurons whose change of activation across two timesteps is below a defined threshold. This work describes a training algorithm for Delta RNNs that exploits temporal sparsity in the backward propagation phase to reduce computational requirements for training on the edge. Due to the symmetric computation graphs of forward and backward propagation during training, the gradient computation of inactivated neurons can be skipped. Results show a reduction of ∼80% in matrix operations for training a 56k parameter Delta LSTM on the Fluent Speech Commands dataset with negligible accuracy loss. Logic simulations of a hardware accelerator designed for the training algorithm show 2-10X speedup in matrix computations for an activation sparsity range of 50%-90%. Additionally, we show that the proposed Delta RNN training will be useful for online incremental learning on edge devices with limited computing resources.

Dynamic task scheduling in edge cloud systems using deep recurrent neural networks and environment learning approaches

The rapid growth of the cloud computing landscape has created significant challenges in managing the escalating volume of data and diverse resources within the cloud environment, catering to a broad spectrum of users ranging from individuals to large corporations. Ineffectual resource allocation in cloud systems poses a threat to overall performance, necessitating the equitable distribution of resources among stakeholders to ensure profitability and customer satisfaction. This paper addresses the critical issue of resource management in cloud computing through the introduction of a Dynamic Task Scheduling with Virtual Machine allocation (DTS-VM) strategy, incorporating Edge-Cloud computing for the Internet of Things (IoT). The proposed approach begins by employing a Recurrent Neural Network (RNN) algorithm to classify user tasks into Low Priority, Mid Priority, and High Priority categories. Tasks are then assigned to Edge nodes based on their priority, optimizing efficiency through the application of the Spotted Hyena Optimization (SHO) algorithm for selecting the most suitable edge node. To address potential overloads on the edge, a Fuzzy approach evaluates offloading decisions using multiple metrics. Finally, optimal Virtual Machine allocation is achieved through the application of the Stable Matching algorithm. The seamless integration of these components ensures a dynamic and efficient allocation of resources, preventing the prolonged withholding of customer requests due to the absence of essential resources. The proposed system aims to enhance overall cloud system performance and user satisfaction while maintaining organizational profitability. The effectiveness of the DTS-VM strategy is validated through comprehensive testing and evaluation, showcasing its potential to address the challenges posed by the diverse and expanding cloud computing landscape.

Climatic scenario-based integrated recurrent ensemble model for energy demand forecasting

This study presents an improved approach to estimate future energy demand based on climate scenarios, upstream documents, and Machine Learning (ML). To accurately predict future weather conditions, a Regional Climate Model (RCM) is developed. The RCM incorporates an ensemble approach that combines the power of recurrent neural network and recurrent random forest algorithms to analyze relevant data. The demand model incorporates correction coefficients that are derived from behavioral and variations in income, coupled with an ensemble and recurrent ML models. Four scenarios named from the distinctive wildlife are considered for Tehran, I. R. Iran as a real case study for megacities. The average temperature in the Caracal, Striped hyena, and Balochistan black bear scenarios is projected to be 0.91 °C, 2.01 °C, and 2.86 °C higher than the Persian cheetah scenario between 2015 and 2100. Similarly, natural gas demands are estimated to be 0.93, 4.91, and 3.38 times higher than the Persian cheetah scenario by 2050, respectively. The electricity demand is also 14, 3.98, and 3.50 times higher in the Caracal, Striped hyena, and Balochistan black scenarios compared to the Persian cheetah.This study has developed scenarios delineating future temperature conditions, as well as revealing how energy demand changes within climate change. The results of this study precisely identify the influential features for predicting future energy demand and temperature, indicating that utilizing recurrent models is preferable for forecasting future temperature and energy demand.These changes in the energy demand have significant implications for investment in the energy supply sector as well as the greenhouse gas emissions and the air pollution levels. Therefore, it is crucial to prioritize climate scenarios that lead to a decrease in energy demand to reach sustainable zero-emission cities.

Flow starvation during square-flow assisted ventilation detected by supervised deep learning techniques

BackgroundFlow starvation is a type of patient-ventilator asynchrony that occurs when gas delivery does not fully meet the patients’ ventilatory demand due to an insufficient airflow and/or a high inspiratory effort, and it is usually identified by visual inspection of airway pressure waveform. Clinical diagnosis is cumbersome and prone to underdiagnosis, being an opportunity for artificial intelligence. Our objective is to develop a supervised artificial intelligence algorithm for identifying airway pressure deformation during square-flow assisted ventilation and patient-triggered breaths.MethodsMulticenter, observational study. Adult critically ill patients under mechanical ventilation > 24 h on square-flow assisted ventilation were included. As the reference, 5 intensive care experts classified airway pressure deformation severity. Convolutional neural network and recurrent neural network models were trained and evaluated using accuracy, precision, recall and F1 score. In a subgroup of patients with esophageal pressure measurement (ΔPes), we analyzed the association between the intensity of the inspiratory effort and the airway pressure deformation.Results6428 breaths from 28 patients were analyzed, 42% were classified as having normal-mild, 23% moderate, and 34% severe airway pressure deformation. The accuracy of recurrent neural network algorithm and convolutional neural network were 87.9% [87.6–88.3], and 86.8% [86.6–87.4], respectively. Double triggering appeared in 8.8% of breaths, always in the presence of severe airway pressure deformation. The subgroup analysis demonstrated that 74.4% of breaths classified as severe airway pressure deformation had a ΔPes > 10 cmH2O and 37.2% a ΔPes > 15 cmH2O.ConclusionsRecurrent neural network model appears excellent to identify airway pressure deformation due to flow starvation. It could be used as a real-time, 24-h bedside monitoring tool to minimize unrecognized periods of inappropriate patient-ventilator interaction.

Design of signage guidance system for tourist attractions based on computer vision technology

In order to improve tourists’ sense of touring experience, this research proposes a signage navigation system for mobile scenic spots. Firstly, the system uses a recurrent neural network algorithm incorporating convolutional neural network for image recognition function to obtain relevant information through image recognition. Then the target localization of the image is performed according to the single-stage target detection algorithm, and the location of the landscape in which the user is located is localized by the recognized image information. The results show that the algorithm can achieve 86.7% recognition accuracy, and it can recognize part of the image samples when the recognition time reaches the range of 0.8 min–1 min. The single-stage target detection algorithm has a localization accuracy of 97.2% with a minimum loss rate of 1.1%. And the algorithm has high average accuracy and full class average accuracy values. The system has good application value.