Term Memory Recurrent Neural Network Research Articles

Adaptive Anomaly Detection Framework Model Objects in Cyberspace.

Telecommunication has registered strong and rapid growth in the past decade. Accordingly, the monitoring of computers and networks is too complicated for network administrators. Hence, network security represents one of the biggest serious challenges that can be faced by network security communities. Taking into consideration the fact that e-banking, e-commerce, and business data will be shared on the computer network, these data may face a threat from intrusion. The purpose of this research is to propose a methodology that will lead to a high level and sustainable protection against cyberattacks. In particular, an adaptive anomaly detection framework model was developed using deep and machine learning algorithms to manage automatically-configured application-level firewalls. The standard network datasets were used to evaluate the proposed model which is designed for improving the cybersecurity system. The deep learning based on Long-Short Term Memory Recurrent Neural Network (LSTM-RNN) and machine learning algorithms namely Support Vector Machine (SVM), K-Nearest Neighbor (K-NN) algorithms were implemented to classify the Denial-of-Service attack (DoS) and Distributed Denial-of-Service (DDoS) attacks. The information gain method was applied to select the relevant features from the network dataset. These network features were significant to improve the classification algorithm. The system was used to classify DoS and DDoS attacks in four stand datasets namely KDD cup 199, NSL-KDD, ISCX, and ICI-ID2017. The empirical results indicate that the deep learning based on the LSTM-RNN algorithm has obtained the highest accuracy. The proposed system based on the LSTM-RNN algorithm produced the highest testing accuracy rate of 99.51% and 99.91% with respect to KDD Cup'99, NSL-KDD, ISCX, and ICI-Id2017 datasets, respectively. A comparative result analysis between the machine learning algorithms, namely SVM and KNN, and the deep learning algorithms based on the LSTM-RNN model is presented. Finally, it is concluded that the LSTM-RNN model is efficient and effective to improve the cybersecurity system for detecting anomaly-based cybersecurity.

American sign language recognition and training method with recurrent neural network

Though American sign language (ASL) has gained recognition from the American society, few ASL applications have been developed with educational purposes. Those designed with real-time sign recognition systems are also lacking. Leap motion controller facilitates the real-time and accurate recognition of ASL signs. It allows an opportunity for designing a learning application with a real-time sign recognition system that seeks to improve the effectiveness of ASL learning. The project proposes an ASL learning application prototype. The application would be a whack-a-mole game with a real-time sign recognition system embedded. Since both static and dynamic signs (J, Z) exist in ASL alphabets, Long-Short Term Memory Recurrent Neural Network with k-Nearest-Neighbour method is adopted as the classification method is based on handling of sequences of input. Characteristics such as sphere radius, angles between fingers and distance between finger positions are extracted as input for the classification model. The model is trained with 2600 samples, 100 samples taken for each alphabet. The experimental results revealed that the recognition rate for 26 ASL alphabets yields an average of 99.44% accuracy rate and 91.82% in 5-fold cross-validation with the use of leap motion controller.

Predicting PTSD severity using longitudinal magnetoencephalography with a multi-step learning framework

Objective. The present study explores the effectiveness of incorporating temporal information in predicting post-traumatic stress disorder (PTSD) severity using magnetoencephalography (MEG) imaging data. The main objective was to assess the relationship between longitudinal MEG functional connectome data, measured across a variety of neural oscillatory frequencies and collected at two timepoints (Phase I and II), against PTSD severity captured at the later time point. Approach. We used an in-house developed informatics solution, featuring a two-step process featuring pre-learn feature selection (CV-SVR-rRF-FS, cross-validation with support vector regression (SVR) and recursive random forest feature selection) and deep learning (long-short term memory recurrent neural network, LSTM-RNN) techniques. Main results. The pre-learn step selected a small number of functional connections (or edges) from Phase I MEG data associated with Phase II PTSD severity, indexed using the PTSD CheckList (PCL) score. This strategy identified the functional edges affected by traumatic exposure and indexed disease severity, either permanently or evolving dynamically over time, for optimal predictive performance. Using the selected functional edges, LSTM modelling was used to incorporate the Phase II MEG data into longitudinal regression models. Single timepoint (Phase I and Phase II MEG data) SVR models were generated for comparison. Assessed with holdout test data, alpha and high gamma bands showed enhanced predictive performance with the longitudinal models comparing to the Phase I single timepoint models. The best predictive performance was observed for lower frequency ranges compared to the higher frequencies (low gamma), for both model types. Significance. This study identified the neural oscillatory signatures that benefited from additional temporal information when estimating the outcome of PTSD severity using MEG functional connectome data. Crucially, this approach can similarly be applied to any other mental health challenge, using this effective informatics foundation for longitudinal tracking of pathological brain states and predicting outcome with a MEG-based neurophysiology imaging system.

Human context recognition with reduced feature space vectors for resource constraint gadgets

The aim of this paper is to determine the minimal features to train classification algorithms pertaining to context recognition using data collected from accelerometer sensor in smartphones. A detailed experimental evaluation of various time and frequency domain features on raw accelerometer sensor data collected from smartphones leads to the most influential minimal collection that aid in recognizing human context in resource constraint gadgets. To substantiate the study, four classifiers, namely, Logistic Regression, Support Vector Machine, Artificial Neural Network and Long Short Term Memory Recurrent Neural Networks are trained on six activities - Sitting, Standing and lying (sedentary activities), Walking, Walking Upstairs and Walking Downstairs (dynamic activities). The raw accelerometer values from UCI public dataset and the data collected from Android application is used to build classification model. Classifier performance on both datasets showed 90% accuracy with four features taken over each axes of 3-axis accelerometer.

Study on Constitutive Relation of Nickel-Base Superalloy Inconel 718 Based on Long Short Term Memory Recurrent Neural Network

The high temperature tensile test of Inconel 718 under the conditions of deformation temperature of 950 °C–1100 °C and strain rate of 0.0005 s−1–0.1 s−1 was carried out, and its true stress–true strain curve was drawn. Through the analysis of the flow stress of Inconel 718 under different conditions, it can be seen that the high-temperature rheological behavior of Inconel 718 is affected by the coupling of strain hardening effect and dynamic softening effect, and has significant loading history correlation. By applying the stretched data, a long short term memory (LSTM) recurrent neural network was trained to characterize the constitutive relationship of Inconel 718. The experimental results show that the prediction results of the LSTM constitutive model are extremely consistent with the experimental data, which is significantly better than the modified Johnson–Cook (M-JC) model. Finally, high temperature tensile experiments under variable strain rates were carried out to verify the feasibility of the LSTM constitutive model in the complex loading and unloading stages.

A Multi-Institutional External Validation of a Deep-Learning Based Platform for Prediction of Outcomes following SBRT Treatment for Early-Stage Non-Small Cell Lung Cancer

Despite favorable tumor control of NSCLC with SBRT, some patients remain at high risk for treatment failure. Our group previously developed a deep learning (DL) platform that integrated clinical, radiographic, and dosimetric data to predict outcomes in patients with early-stage NSCLC at a single institution. This study aimed to externally validate the generalizability of DL based predictions on a blinded cohort of patients at a geographically distinct institution and assess model performance versus traditional statistical models. A DL algorithm was trained on a cohort of 527 patients with cT1-cT2 NSCLC treated with SBRT at a single-institution. The DL algorithm incorporated feed-forward neural network architecture to model clinical co-variates, 2D convolutional neural network (CNN) and long-short term memory recurrent neural network architectures to model radiographic data, and 1D CNN to model dosimetric data. Outcomes of interest were all-cause mortality as well as local, regional, distant, and overall treatment failures at 2-years. The DL model was tested with a blinded cohort of 62 patients treated at an external institution. Discriminatory ability of the DL model on the validation set was assessed using receiver operating characteristic curves (ROC) and compared to predictions generated from a traditional Cox Proportional Hazard model. Statistically significant differences in ROC curves were tested using the Delong method. The training and blinded validation cohorts showed no statistically significant differences in disease presentation, workup, or outcomes. When tested on the blinded external validation cohort, the DL platform maintained strong discriminatory ability in predicting all-cause mortality (AUC 0.73) as well as local (AUC 0.79), regional (AUC 0.69), distant (AUC 0.72), and overall (AUC 0.85) treatment failures. Discriminatory ability of the DL platform significantly outperformed the predictions generated from a traditional Cox Proportional Hazard model for all outcomes of interest (Table 1). A DL platform that effectively integrates clinical, imaging, and dosimetric data successfully models outcomes in early-stage NSCLC patients treated with SBRT on a blinded external validation cohort. The DL platform shows strong generalizability across multiple institutions and consistently outperforms predictions generated from a traditional Cox Proportional Hazard model. Integration of similar DL models that effectively analyze multiple data streams may improve clinical outcome prediction within clinical practice and guide personalization of therapy.Abstract 1103; Table2-Year OutcomesAUC DL Model95% Confidence IntervalAUC Cox Model95% Confidence IntervalDelong TestAny Treatment Failure.85.70-1.00.65.52-.78p<.001Local Failure.79.62-.96.58.47-.69p<.001Distant Failure.72.56-.88.60.47-.73p = .008Regional Failure.69.52-.86.61.51-.71p = .041All-Cause Mortality.73.61-.85.60.48-.72p<.001 Open table in a new tab

Clinical Context-Aware Biomedical Text Summarization Using Deep Neural Network: Model Development and Validation.

BackgroundAutomatic text summarization (ATS) enables users to retrieve meaningful evidence from big data of biomedical repositories to make complex clinical decisions. Deep neural and recurrent networks outperform traditional machine-learning techniques in areas of natural language processing and computer vision; however, they are yet to be explored in the ATS domain, particularly for medical text summarization.ObjectiveTraditional approaches in ATS for biomedical text suffer from fundamental issues such as an inability to capture clinical context, quality of evidence, and purpose-driven selection of passages for the summary. We aimed to circumvent these limitations through achieving precise, succinct, and coherent information extraction from credible published biomedical resources, and to construct a simplified summary containing the most informative content that can offer a review particular to clinical needs.MethodsIn our proposed approach, we introduce a novel framework, termed Biomed-Summarizer, that provides quality-aware Patient/Problem, Intervention, Comparison, and Outcome (PICO)-based intelligent and context-enabled summarization of biomedical text. Biomed-Summarizer integrates the prognosis quality recognition model with a clinical context–aware model to locate text sequences in the body of a biomedical article for use in the final summary. First, we developed a deep neural network binary classifier for quality recognition to acquire scientifically sound studies and filter out others. Second, we developed a bidirectional long-short term memory recurrent neural network as a clinical context–aware classifier, which was trained on semantically enriched features generated using a word-embedding tokenizer for identification of meaningful sentences representing PICO text sequences. Third, we calculated the similarity between query and PICO text sequences using Jaccard similarity with semantic enrichments, where the semantic enrichments are obtained using medical ontologies. Last, we generated a representative summary from the high-scoring PICO sequences aggregated by study type, publication credibility, and freshness score.ResultsEvaluation of the prognosis quality recognition model using a large dataset of biomedical literature related to intracranial aneurysm showed an accuracy of 95.41% (2562/2686) in terms of recognizing quality articles. The clinical context–aware multiclass classifier outperformed the traditional machine-learning algorithms, including support vector machine, gradient boosted tree, linear regression, K-nearest neighbor, and naïve Bayes, by achieving 93% (16127/17341) accuracy for classifying five categories: aim, population, intervention, results, and outcome. The semantic similarity algorithm achieved a significant Pearson correlation coefficient of 0.61 (0-1 scale) on a well-known BIOSSES dataset (with 100 pair sentences) after semantic enrichment, representing an improvement of 8.9% over baseline Jaccard similarity. Finally, we found a highly positive correlation among the evaluations performed by three domain experts concerning different metrics, suggesting that the automated summarization is satisfactory.ConclusionsBy employing the proposed method Biomed-Summarizer, high accuracy in ATS was achieved, enabling seamless curation of research evidence from the biomedical literature to use for clinical decision-making.

Multi-objective long-short term memory recurrent neural networks for speech enhancement

Speech-in-noise perception is an important research problem in many real-world multimedia applications. The noise-reduction methods contributed significantly; however rely on a priori information about the noise signals. Deep learning approaches are developed for enhancing the speech signals in nonstationary noisy backgrounds and their benefits are evaluated for the perceived speech quality and intelligibility. In this paper, a multi-objective speech enhancement based on the Long-Short Term Memory (LSTM) recurrent neural network (RNN) is proposed to simultaneously estimate the magnitude and phase spectra of clean speech. During training, the noisy phase spectrum is incorporated as a target and the unstructured phase spectrum is transformed to its derivative that has an identical structure to corresponding magnitude spectrum. Critical Band Importance Functions (CBIFs) are used in training process to further improve the network performance. The results verified that the proposed multi-objective LSTM (MO-LSTM) successfully outscored the standard magnitude-aware LSTM (MA-LSTM), magnitude-aware DNN (MA-DNN), phase-aware DNN (PA-DNN), magnitude-aware GNN (MA-GNN) and magnitude-aware CNN (MA-CNN). Moreover, the proposed speech enhancement considerably improved the speech quality, intelligibility, noise-reduction and automatic speech recognition in changing noisy backgrounds, which is confirmed by the ANalysis Of VAriance (ANOVA) statistical analysis.

Predicting irregularities in arrival times for transit buses with recurrent neural networks using GPS coordinates and weather data

Intelligent transportation systems (ITS) play an important role in the quality of life of citizens in any metropolitan city. Despite various policies and strategies incorporated to increase the reliability and quality of service, public transportation authorities continue to face criticism from commuters largely due to irregularities in bus arrival times, most notably manifested in early or late arrivals. Due to these irregularities, commuters may miss important appointments, wait for too long at the bus stop, or arrive late for work. Therefore, accurate prediction models are needed to build better customer service solutions for transit systems, e.g. building accurate mobile apps for trip planning or sending bus delay/cancel notifications. Prediction models will also help in developing better appointment scheduling systems for doctors, dentists, and other businesses to take into account transit bus delays for their clients. In this paper, we seek to predict the occurrence of arrival time irregularities by mining GPS coordinates of transit buses provided by the Toronto Transit Commission (TTC) along with hourly weather data and using this data in machine learning models that we have developed. In our study, we compared the performance of a Long Short Term Memory Recurrent Neural Network (LSTM) model against four baseline models, an Artificial Neural Network (ANN), Support Vector Regression (SVR), Autoregressive Integrated Moving Average (ARIMA) and historical averages. We found that our LSTM model demonstrates the best prediction accuracy. The improved accuracy achieved by the LSTM model may lend itself to its ability to adjust and update the weights of neurons while accounting for long-term dependencies. In addition, we found that weather conditions play a significant role in improving the accuracy of our models. Therefore, we built a prediction model that combines an LSTM model with a Recurrent Neural Network Model (RNN) that focuses on the weather condition. Our findings also reveal that in nearly 37% of scheduled arrival times, buses either arrive early or late by a margin of more than 5 min, suggesting room for improvement in the current strategies employed by transit authorities.

Realtime mobile bandwidth prediction using LSTM neural network and Bayesian fusion

With the increasing popularity of mobile Internet and the higher bandwidth requirement of mobile applications, user Quality of Experience (QoE) is particularly important. For applications requiring high bandwidth and low delay, such as video streaming, video conferencing, and online gaming, etc., if the future bandwidth can be estimated in advance, applications can leverage the estimation to adjust their data transmission strategies and significantly improve the user QoE. In this paper, we focus on accurate bandwidth prediction to improve user QoE. Specifically, We study realtime mobile bandwidth prediction in various mobile networking scenarios, such as subway and bus rides along different routes. The primary method used is Long Short Term Memory (LSTM) recurrent neural network. In individual scenarios, LSTM significantly improves the prediction accuracy of state-of-the-art prediction algorithms, such as Recursive Least Squares (RLS) by 12% in Root Mean Square Error (RMSE) and by 17% in Mean Average Error (MAE). We further developed Multi-Scale Entropy (MSE) to analyze the bandwidth patterns in different mobility scenarios and discuss its connection to the achieved accuracy. For practical applications, we developed Model Switching and Bayes Model Fusion to use pre-trained LSTM models for online realtime bandwidth prediction.

Distracted driver detection by combining in-vehicle and image data using deep learning

Distracted driving is among the most important reasons for traffic accidents today. Recently, there is an increasing interest in building driver assistance systems that detect the actions of the drivers and help them drive safer. In these studies, although some distinct data types such as the physical conditions of the driver, audio and visual features, car information are used; the main data source is the images of the driver that include the face, arms, and hands taken with a camera placed inside the car. In this work, we propose to integrate sensor data into the vision-based distracted driver detection model to improve the generalization ability of the system. With this purpose, we created a new data set that includes driver images and sensor data collected from real-world drives. Then, we constructed a two-stage distracted driving detection system to detect nine distracted behaviors. In the first stage, vision-based Convolutional Neural Network (CNN) models were created by transfer learning and fine-tuning methods. In the second stage, Long-Short Term Memory-Recurrent Neural Network (LSTM-RNN) models were created using sensor and image data together. We evaluate our system by two different fusion techniques and show that integrating sensor data to image-based driver detection significantly increases the overall performance with both of the fusion techniques. We also show that the accuracy of the vision-based model increases by fine-tuning the pre-trained CNN model using a related public dataset.

Toward Packet Routing With Fully Distributed Multiagent Deep Reinforcement Learning

Packet routing is one of the fundamental problems in computer networks in which a router determines the next-hop of each packet in the queue to get it as quickly as possible to its destination. Reinforcement learning (RL) has been introduced to design autonomous packet routing policies with local information of stochastic packet arrival and service. However, the curse of dimensionality of RL prohibits the more comprehensive representation of dynamic network states, thus limiting its potential benefit. In this article, we propose a novel packet routing framework based on <i>multiagent</i> deep RL (DRL) in which each router possess an <i>independent</i> long short term memory (LSTM) recurrent neural network (RNN) for training and decision making in a <i>fully distributed</i> environment. The LSTM RNN extracts routing features from rich information regarding backlogged packets and past actions, and effectively approximates the value function of Q-learning. We further allow each route to communicate periodically with direct neighbors so that a broader view of network state can be incorporated. The experimental results manifest that our multiagent DRL policy can strike the delicate balance between congestion-aware and shortest routes, and significantly reduce the packet delivery time in general network topologies compared with its counterparts.

RETRACTED: Forecasting biofuel production using adaptive integrated optimization network model

Abstract Recently, biofuel production has been increasing all over the world over the past few decades owing to the relative changes in energy price and policy intervention. Several research studies are focussed on producing optimal biofuel production using numerous approaches. However, those approaches have few drawbacks such as computational complexity, less accuracy, high error value rate, etc. So to overcome such shortcomings, this paper aims in developing a novel Adaptive Integrated Optimization Network (AION approach) to attain optimal biofuel production with high accuracy and minimum error value rate. Moreover, the proposed AION approach comprises four significant phases namely Pre-processing of data, Re-construction of component, Prediction of an individual and Ensemble forecasting. The initial process utilizes an empirical mode decomposition (EMD) approach to divide a raw complicated data. The reconstruction phase employs the Fine to coarse (FTC) reconstructing approach that rebuilds the IMF into two different t-testing components namely the low-frequency component and the high-frequency component. Moreover, Gray Wolf Optimization (GWO) based Adaptive Wavelet Neural Network (AWNN) and Long Short Term Memory-Recurrent Neural Network (LSTM-RNN) approaches are applied to predict the low and the high-frequency components. Then the forecasted results are summed by using an ADD operator so as to obtain absolute forecasting results. The experimental analysis reveals that the proposed AION approach provides an optimal biofuel production with enhanced accuracy, minimum error value rate, and complexity issues.

A Novel Automated Blood Pressure Estimation Algorithm Using Sequences of Korotkoff Sounds.

The use of automated non-invasive blood pressure (NIBP) measurement devices is growing, as they can be used without expertise, and BP measurement can be performed by patients at home. Non-invasive cuff-based monitoring is the dominant method for BP measurement. While the oscillometric technique is most common, a few automated NIBP measurement methods have been developed based on the auscultatory technique. Amongst artificial intelligence (AI) techniques, deep learning has received increasing attention in different fields due to its strength in data classification, and feature extraction problems. This paper proposes a novel automated AI-based technique for NIBP estimation from auscultatory waveforms (AWs) based on converting the NIBP estimation problem to a sequence-to-sequence classification problem. To do this, a sequence of segments was first formed by segmenting the AWs, and their corresponding decomposed detail, and approximation parts obtained by wavelet packet decomposition method, and extracting features from each segment. Then, a label was assigned to each segment, i.e. (i) between systolic, and diastolic segments, and (ii) otherwise, and a bidirectional long short term memory recurrent neural network (BiLSTM-RNN) was devised to solve the resulting sequence-to-sequence classification problem. Adopting a 5-fold cross-validation scheme, and using a data base of 350 NIBP recordings gave an average mean absolute error of 1.7±3.7 mmHg for systolic BP (SBP), and 3.4 ±5.0 mmHg for diastolic BP (DBP) relative to reference values. Based on the results achieved, and comparisons made with the existing literature, it is concluded that the proposed automated BP estimation algorithm based on deep learning methods, and auscultatory waveform brings plausible benefits to the field of BP estimation.

Prognostic for fuel cell based on particle filter and recurrent neural network fusion structure

Fuel cells are considered as one of the most promising candidates for future power source due to its high energy density and environmentally friendly properties, whereas the short lifespan blocks its large-scale commercialization. In order to enhance the reliability and durability of proton exchange membrane fuel cell, a fusion prognostic approach based on particle filter (model-based) and long-short term memory recurrent neural network (data-driven) is proposed in this paper. Both the remaining useful life estimation and the short-term degradation prediction can be achieved based on the prognostic method. For remaining useful life estimation, the particle filter method is used to identify the model parameters in the training phase and the long-short term memory recurrent neural network is used to update the parameters in the prediction phase. As for short-term degradation prediction, the particle filter and long-short term memory recurrent neural network are firstly trained individually in the training phase and then be fused to make predictions in the prediction phase. The proposed fusion structure is validated by the fuel cell experimental tests data, and results indicate that better prognostic performance can be obtained compared with the individual model-based or data-driven method.

Text-independent speaker recognition using LSTM-RNN and speech enhancement

Speaker recognition revolution has lead to the inclusion of speaker recognition modules in several commercial products. Most published algorithms for speaker recognition focus on text-dependent speaker recognition. In contrast, text-independent speaker recognition is more advantageous as the client can talk freely to the system. In this paper, text-independent speaker recognition is considered in the presence of some degradation effects such as noise and reverberation. Mel-Frequency Cepstral Coefficients (MFCCs), spectrum and log-spectrum are used for feature extraction from the speech signals. These features are processed with the Long-Short Term Memory Recurrent Neural Network (LSTM-RNN) as a classification tool to complete the speaker recognition task. The network learns to recognize the speakers efficiently in a text-independent manner, when the recording circumstances are the same. The recognition rate reaches 95.33% using MFCCs, while it is increased to 98.7% when using spectrum or log-spectrum. However, the system has some challenges to recognize speakers from different recording environments. Hence, different speech enhancement techniques, such as spectral subtraction and wavelet denoising, are used to improve the recognition performance to some extent. The proposed approach shows superiority, when compared to the algorithm of R. Togneri and D. Pullella (2011).

Attention-Based SeriesNet: An Attention-Based Hybrid Neural Network Model for Conditional Time Series Forecasting

Traditional time series forecasting techniques can not extract good enough sequence data features, and their accuracies are limited. The deep learning structure SeriesNet is an advanced method, which adopts hybrid neural networks, including dilated causal convolutional neural network (DC-CNN) and Long-short term memory recurrent neural network (LSTM-RNN), to learn multi-range and multi-level features from multi-conditional time series with higher accuracy. However, they didn’t consider the attention mechanisms to learn temporal features. Besides, the conditioning method for CNN and RNN is not specific, and the number of parameters in each layer is tremendous. This paper proposes the conditioning method for two types of neural networks, and respectively uses the gated recurrent unit network (GRU) and the dilated depthwise separable temporal convolutional networks (DDSTCNs) instead of LSTM and DC-CNN for reducing the parameters. Furthermore, this paper presents the lightweight RNN-based hidden state attention module (HSAM) combined with the proposed CNN-based convolutional block attention module (CBAM) for time series forecasting. Experimental results show our model is superior to other models from the viewpoint of forecasting accuracy and computation efficiency.

Retracted: Using Fuzzy Uncertainty Quantization and Hybrid RNN-LSTM Deep Learning Model for Wind Turbine Power

Along with the many benefits, which the high penetration of wind turbine has brought to the smart grids, the issues arising due to the uncertainty effect is an inevitable phenomenon, which needs to be controlled. The high uncertainty effects can cause unstable operation and management scheduling plans, which can cause severe problems for both the operator and customers. This paper proposes a new quantization approach based on hybrid deep learning to capture the forecast error in the wind turbine output power. The proposed method is equipped with a long-short term memory (LSTM) model and recurrent neural network (RNN) to learn the most effective spatial-temporal features in wind turbine output power. Due to the high complexity of the data, a new optimization method based on a modified sine cos algorithm is proposed to help more stable training of the model. The feasibility and appropriate performance of the model is assessed through experimental analysis on two datasets in the Australia wind farms.

Cross Entropy Based Long Short Term Memory Recurrent Neural Network Model for Analyzing the Time Series on Stock Market Price

Cross Entropy Based Long Short Term Memory Recurrent Neural Network Model for Analyzing the Time Series on Stock Market Price

Smart Detection of Fire Source in Tunnel Based on the Numerical Database and Artificial Intelligence

The fire event in a tunnel creates a rapid spread of heat and smoke flows in a long and confined space, which not only endangers human life but also challenges the fire-evacuation and firefighting strategies. A quick and accurate identification for the location and size of the original fire source is of great scientific and practical value in guiding fire rescue and fighting the tunnel fire. Nevertheless, it is a big challenge to acquire fire-source information in an actual tunnel fire event. In this study, the framework of artificial intelligence (AI) and big data is applied to predict the fire source in a numerical model of the tunnel. A big tunnel fire database of numerical simulations, with varying fire locations, fire sizes, and ventilation conditions, is constructed. Temporally varied temperatures measured by multiple sensor devices are used to train a long-short term memory recurrent neural network. Results demonstrate that the location and size of the tunnel fire and the ventilation wind speed can be predicted by the trained model with an accuracy of 90%. Sensitivity analysis is also carried out to optimize the database configuration and spatial–temporal arrangement of sensors in order to achieve a fast and reliable fire prediction. This work addresses the possibility of AI-based detection and prediction of fire source and hazard, thus, providing scientifically based guidance for smart-firefighting technologies and paving the way for future emergency-response tactics in a smart city.