Long Short-term Memory Model Research Articles

TRANSFORMING BANKING SECURITY: THE ROLE OF DEEP LEARNING IN FRAUD DETECTION SYSTEMS

In the digital banking landscape, the increasing volume of online transactions has heightened the risk of fraudulent activities, necessitating the development of more effective detection systems. This study investigates the efficacy of various machine learning and deep learning algorithms in identifying fraudulent transactions, emphasizing Long Short-Term Memory (LSTM) networks. We implemented and evaluated multiple algorithms, including Logistic Regression, Random Forest, Gradient Boosting Machines (GBM), and XGBoost, on a large-scale credit card transaction dataset. Our results demonstrate that the LSTM model outperforms traditional machine learning algorithms, achieving an accuracy of 98.5%, precision of 87.2%, recall of 85.0%, and an Area Under the Curve (AUC) score of 0.94. These findings highlight the superior capability of LSTM networks to capture complex patterns in sequential transaction data, making them an asset for real-time fraud detection in banking. This research underscores the need for financial institutions to adopt advanced deep learning techniques to enhance their fraud detection systems, thereby minimizing financial losses and improving customer trust.

An improved prediction method for BDS-3 SISA parameters and the preliminary performance evaluation

Abstract The ephemeris accuracy parameters broadcast in the ephemeris messages are commonly utilized as an alarm threshold for the signal-in-space ranging errors (SISRE) of the Global Navigation Satellite System (GNSS). Accurately predicting the orbit and clock errors of broadcast ephemeris, and converting the prediction errors into broadcast ephemeris accuracy parameters using a robust model, is important for the integrity service of GNSS. The BeiDou Navigation Satellite System (BDS-3) broadcasts four ephemeris accuracy parameters, namely SIS A oe , SIS A ocb , SIS A oc1 , and SIS A oc2 , to represent the standard deviations of the satellite orbit and clock prediction errors in the broadcast ephemeris. The probability that the broadcast ephemeris SISRE exceeds 4.42 times the signal-in-space accuracy (SISA) without triggering an alarm in time is expected to be less than 1 × 10−5. In addition to accurately enveloping the broadcast ephemeris SISRE, SISA should also reduce the over-envelopment margin over SISRE to accurately express the system service accuracy. Thus, an accurate prediction of orbit and clock errors in the broadcast ephemeris is crucial for calculating the SISA. An orbit and clock error prediction method based on the Long Short-Term Memory (LSTM) neural network is proposed in this study. This method uses historical satellite orbit and clock errors, derived from the difference between precise and broadcast ephemeris, to train an LSTM neural network. The trained LSTM model is then used to predict the orbit and clock errors for the next 24 h. The predicted errors are used to calculate the corresponding SISA for the broadcast ephemeris. The results indicate that the 24 h accuracies of the predicted orbit and clock errors are less than 0.06 m and 0.25 m, respectively. The calculated SISA can completely envelop the actual broadcast ephemeris SISRE, achieving 100% coverage while also decreasing the over-envelopment margin by 80%. Furthermore, the horizontal protection level and vertical protection level calculated using the corresponding SISA are reduced by approximately 14%, meeting the integrity requirements set by the International Civil Aviation Organization for Approach with Vertical Guidance I. The proposed method enhances the timeliness of SISA parameter calculation and the integrity and availability of the BDS-3 service, serving as a crucial reference value for safety critical applications such as civil aviation services.

A groundwater level spatiotemporal prediction model based on graph convolutional networks with a long short-term memory

ABSTRACT The performance of regional groundwater level (GWL) prediction model hinges on understanding intricate spatiotemporal correlations among monitoring wells. In this study, a graph convolutional network (GCN) with a long short-term memory (LSTM) (GCN–LSTM) model is introduced for GWL prediction utilizing data from 16 wells located in the northeastern Xiangtan City, China. This model is designed to account for both the hybrid temporal dependencies and spatial autocorrelations among wells. It consists of two parts: the spatial part employs GCNs to extract spatial characteristics from a spatial self-similarity weight matrix and an attribute self-similarity wight matrix among wells; the temporal part utilizes a LSTM module to capture the temporal patterns of GWL sequences, along with monthly precipitation and temperature data. This model dynamically predicts changes in groundwater levels, achieving higher accuracy on average compared to single-well predictions using LSTM. By incorporating both temporal dependencies and spatial autocorrelations, the GCN–LSTM model demonstrated an average improvement in goodness-of-fit of approximately 11.21% over the LSTM-based model for individual wells. Its application holds significant reference value for the sustainable utilization and development of groundwater resources in Xiangtan City.

Prediction and assessment of meteorological drought in southwest China using long short-term memory model

Abstract Drought prediction is crucial for mitigating risks and designing measures to alleviate its impact. Machine learning models have been widely applied in the field of drought prediction in recent years. This study concentrated on predicting meteorological droughts in southwest China, a region prone to frequent and severe droughts, particularly in areas with sparse meteorological station coverage. The long short-term memory (LSTM) predictive model, which is a deep learning model, was constructed by calculating standardized precipitation evapotranspiration index (SPEI) values based on 144 weather station observations from 1980 to 2020. The 5-fold cross-validation method was used for the hyperparameter optimization of the model. The LSTM model underwent comprehensive assessment and validation through multiple methods. This included the use of several accuracy assessment indicators and a comparison of results. The comparison covered different drought characteristics among the LSTM predictive model, the benchmark random forest (RF) predictive model, the historical drought situations, and the calculated SPEI values based on observations from 144 weather stations. The results showed that the training results of the LSTM predictive model basically agreed with the SPEI values calculated from weather station observations. The model-predicted variation trend of SPEI values for 2020 was similar to the variation in SPEI values calculated based on weather station observations. On the test set, the coefficient of determination (R 2), the root mean square error, the explained variance score, the Nash–Sutcliffe efficiency, and the Kling–Gupta efficiency were 0.757, 0.210, 0.802, 0.761, and 0.212, respectively. The total consistency rate of the drought grade was 59.26%. The spatial correlation distribution of SPEI values between LSTM model prediction and calculation from meteorological stations in 2020 was more than 0.5 for most regions. The correlation coefficients exceeded 0.6 in western Tibet and Chengdu Plains. Compared to the RF model, the LSTM model excelled in all five performance evaluation metrics and demonstrated a higher overall consistency rate for drought categories. The Kruskal–Wallis test for both the LSTM and RF models all indicated no significant difference in the distributions between the predicted and observed data. Scatter plots revealed that the prediction accuracy for both models in 2020 was suboptimal, with the SPEI showing a comparatively narrow range of values. Nonetheless, the LSTM model significantly outperformed the RF model in terms of prediction accuracy. In summary, the LSTM model demonstrated good overall performance, accuracy, and applicability. It has the potential to enhance dynamic drought prediction in regions with complex terrain, diverse climatic factors, and sparse weather station networks.

An Efficient Deep Learning Framework for Optimized Event Forecasting

There have been several catastrophic events that have impacted multiple economies and resulted in thousands of fatalities, and violence has generated a severe political and financial crisis. Multiple studies have been centered around the artificial intelligence (AI) and machine learning (ML) approaches that are most widely used in practice to detect or forecast violent activities. However, machine learning algorithms become less accurate in identifying and forecasting violent activity as data volume and complexity increase. For the prediction of future events, we propose a hybrid deep learning (DL)-based model that is composed of a convolutional neural network (CNN), long short-term memory (LSTM), and an attention layer to learn temporal features from the benchmark the Global Terrorism Database (GTD). The GTD is an internationally recognized database that includes around 190,000 violent events and occurrences worldwide from 1970 to 2020. We took into account two factors for this experimental work: the type of event and the type of object used. The LSTM model takes these complex feature extractions from the CNN first to determine the chronological link between data points, whereas the attention model is used for the time series prediction of an event. The results show that the proposed model achieved good accuracies for both cases—type of event and type of object—compared to benchmark studies using the same dataset (98.1% and 97.6%, respectively).

A novel data-driven rolling horizon production planning approach for the plastic industry under the uncertainty of demand and recycling rate

Efficient production planning in the plastic industry requires integrating sustainability practices, particularly the use of recycled plastics. Recycling helps reduce environmental impacts and conserve resources by decreasing the reliance on virgin materials. To develop a responsive plan, dynamic approaches are beneficial to confront fluctuations in uncertain parameters such as recycling rate and market demand. The present study proposes a novel Data-driven Rolling Horizon Planning (DRHP) approach for a sustainable and dynamic production plan in the plastic industry. A dynamic Rolling Horizon (RH) planning framework is formulated as a multi-product, multi-period Mixed Integer Linear Programming (MILP) model for the problem. This model aims to minimize total production cost while taking into account the system’s constraints and sustainability considerations. To deal with uncertainty, the RH-based MILP model is coupled with a rolling Long Short-Term Memory (LSTM) model. The rolling LSTM model leverages historical data of market demand and recycling rate to predict their future values and consequently improve the responsiveness of the production plan. The outperformance of the proposed DRHP approach is demonstrated through an extensive comparison with a robust static counterpart in terms of total production cost and sustainability performance. Results indicate significant reductions, up to 80%, in production costs using the proposed DRHP approach. Furthermore, the effect of rolling duration is investigated concerning production cost, backlog, late-order, and inventory level. Findings highlight the potential of the proposed DRHP approach to mitigate inventory- and late-order-related challenges.

Estimating Summer Maize Biomass by Integrating UAV Multispectral Imagery with Crop Physiological Parameters.

The aboveground biomass (AGB) of summer maize is an important indicator for assessing crop growth status and predicting yield, playing a significant role in agricultural management and decision-making. Traditional on-site measurements of AGB are limited, due to low efficiency and a lack of spatial information. The development of unmanned aerial vehicle (UAV) technology in agriculture offers a rapid and cost-effective method for obtaining crop growth information, but currently, the prediction accuracy of summer maize AGB based on UAVs is limited. This study focuses on the entire growth period of summer maize. Multispectral images of six key growth stages of maize were captured using a DJI Phantom 4 Pro, and color indices and elevation data (DEM) were extracted from these growth stage images. Combining measured data such as summer maize AGB and plant height, which were collected on the ground, and based on the three machine learning algorithms of partial least squares regression (PLSR), random forest (RF), and long short-term memory (LSTM), an input feature analysis of PH was carried out, and a prediction model of summer maize AGB was constructed. The results show that: (1) using unmanned aerial vehicle spectral data (CIS) alone to predict the biomass of summer maize has relatively poor prediction accuracy. Among the three models, the LSTM (CIS) model has the best simulation effect, with a coefficient of determination (R2) ranging from 0.516 to 0.649. The R2 of the RF (CIS) model is 0.446-0.537. The R2 of the PLSR (CIS) model is 0.323-0.401. (2) After adding plant height (PH) data, the accuracy and stability of model prediction significantly improved. R2 increased by about 25%, and both RMSE and NRSME decreased by about 20%. Among the three prediction models, the LSTM (PH + CIS) model had the best performance, with R2 = 0.744, root mean square error (RSME) = 4.833 g, and normalized root mean square error (NRSME) = 0.107. Compared to using only color indices (CIS) as the model input, adding plant height (PH) significantly enhances the prediction effect of AGB (aboveground biomass) prediction in key growth periods of summer maize. This method can serve as a reference for the precise monitoring of crop biomass status through remote sensing with unmanned aerial vehicles.

An Adaptive Combined Method for Lithium‐Ion Battery State of Charge Estimation Using Long Short‐Term Memory Network and Unscented Kalman Filter Considering Battery Aging

AbstractThe accurate estimation of battery state of charge (SOC) enables the reliable and safe operation of lithium‐ion batteries. Data‐driven SOC estimation is considered an emerging and effective solution. However, existing data‐driven SOC estimation methods typically involve direct estimation and lack effective feedback correction. Moreover, battery degradation poses additional challenges to accurate SOC estimation. Therefore, this study proposes an adaptive combined method for battery SOC estimation based on a long short‐term memory (LSTM) network and unscented Kalman filter (UKF) algorithm considering battery aging status. First, an LSTM model is constructed to characterize the battery's dynamic performance instead of traditional battery models. Then, the UKF algorithm is employed to perform SOC estimation through the feedback of terminal voltage prediction. To enhance estimation accuracy under different aging statuses, a proportional‐integral‐derivative controller is employed to correct the capacity fading during the SOC estimation process. Validation results indicate that the terminal voltage prediction model demonstrates exceptional robustness against interference from current and voltage noise. Compared to the traditional estimation method combining the deep learning model and Kalman filter algorithm, the proposed method demonstrates superior estimation accuracy under various complex operating conditions. Furthermore, the proposed method outperforms the traditional method in estimation performance during battery aging.

Non-conventional feature-based LSTM model for prediction of bearing performance degradation

Abstract Mechanical bearings are critical in electrical machines, such as motors and generators. They provide support and facilitate the movement of rotating components in these machines. They are essential to the operation of electrical machinery cause they provide support, reduce friction, ensure proper alignment, and improve overall reliability and efficiency. The appropriate selection, installation, and maintenance of these bearings allow the longevity of such machines. The maintenance and the extent of usage determine the usefulness of the bearings. In predictive maintenance and condition monitoring, accurately predicting the level of bearing degradation holds significant importance for strategizing maintenance tasks, reducing downtime, and avoiding unforeseen breakdowns. This study centers on predicting the level of bearing degradation through the application of neural networks, focusing mainly on non-conventional features, rather than the usual features like kurtosis, and skewness. The dataset utilized is sourced from NASA and includes diverse time series signals that provide insights into the entire lifespan of the bearing. In this investigation, advanced machine learning techniques, long-short term memory (LSTM) in particular, are employed to harness the intricate patterns within the bearing dataset, aiming to improve the accuracy of the predictions.

Prediction of power conversion efficiency parameter of inverted organic solar cells using artificial intelligence techniques

Organic photovoltaic (OPV) cells are at the forefront of sustainable energy generation due to their lightness, flexibility, and low production costs. These characteristics make OPVs a promising solution for achieving sustainable development goals. However, predicting their lifetime remains challenging task due to complex interactions between internal factors such as material degradation, interface stability, and morphological changes, and external factors like environmental conditions, mechanical stress, and encapsulation quality. In this study, we propose a machine learning-based technique to predict the degradation over time of OPVs. Specifically, we employ multi-layer perceptron (MLP) and long short-term memory (LSTM) neural networks to predict the power conversion efficiency (PCE) of inverted organic solar cells (iOSCs) made from the blend PTB7-Th:PC70BM, with PFN as the electron transport layer (ETL), fabricated under an N2 environment. We evaluate the performance of the proposed technique using several statistical metrics, including mean squared error (MSE), root mean squared error (rMSE), relative squared error (RSE), relative absolute error (RAE), and the correlation coefficient (R). The results demonstrate the high accuracy of our proposed technique, evidenced by the minimal error between predicted and experimentally measured PCE values: 0.0325 for RSE, 0.0729 for RAE, 0.2223 for rMSE, and 0.0541 for MSE using the LSTM model. These findings highlight the potential of proposed models in accurately predicting the performance of OPVs, thus contributing to the advancement of sustainable energy technologies.

Comparison of time-series models for predicting physiological metrics under sedation.

This study presents a comprehensive comparison of multiple time-series models applied to physiological metric predictions. It aims to explore the effectiveness of both statistical prediction models and pharmacokinetic-pharmacodynamic prediction model and modern deep learning approaches. Specifically, the study focuses on predicting the bispectral index (BIS), a vital metric in anesthesia used to assess the depth of sedation during surgery, using datasets collected from real-life surgeries. The goal is to evaluate and compare model performance considering both univariate and multivariate schemes. Accurate BIS prediction is essential for avoiding under- or over-sedation, which can lead to adverse outcomes. The study investigates a range of models: The traditional mathematical models include the pharmacokinetic-pharmacodynamic model and statistical models such as autoregressive integrated moving average (ARIMA) and vector autoregression (VAR). The deep learning models encompass recurrent neural networks (RNNs), specifically Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU), as well as Temporal Convolutional Networks (TCNs) and Transformer models. The analysis focuses on evaluating model performance in predicting the BIS using two distinct datasets of physiological metrics collected from actual surgical procedures. It explores both univariate and multivariate prediction schemes and investigates how different combinations of features and input sequence lengths impact model accuracy. The experimental findings reveal significant performance differences among the models: In univariate prediction scenarios for predicting BIS, the LSTM model demonstrates a 2.88% improvement over the second-best performing model. For multivariate predictions, the LSTM model outperforms others by 6.67% compared to the next best model. Furthermore, the addition of Electromyography (EMG) and Mean Arterial Pressure (MAP) brings significant accuracy improvement when predicting BIS. The study emphasizes the importance of selecting and building appropriate time-series models to achieve accurate predictions in biomedical applications. This research provides insights to guide future efforts in improving vital sign prediction methodologies for clinical and research purposes. Clinically, with improvements in the prediction of physiological parameters, clinicians can be informed of interventions if an anomaly is detected or predicted.

Real-time Vehicle Monitoring and Route Prediction using Geolocation through LSTM

Vehicle travel, such as by bus, automobile, etc., is a common way to use public transportation since it is convenient, economical, easily accessible, and eco-friendly. However, there are moments when its challenging to keep an eye on everyone while traveling, avoid traffic bottlenecks, and provide enough service, which makes the passengers bored. In this paper, we offer a real-time geolocation Android application that gathers real-time geolocation data (longitude and latitude), date, time, and speed, and integrates it with Google Firebase to assist predictive analytics for the operational efficiency of a vehicle firm. After being exported from the database as a JSON file, the data must be transformed into a CSV file. The company will be able to anticipate a vehicles future location on a predetermined route based on date, time, and speed by using this database to train an LSTM (Long short-term memory) model, a sort of sequential neural network. Determining the vehicles expected arrival time, allocating passengers, and predicting traffic conditions will therefore be useful in preventing congestion. Utilizing this technology allows a car company to improve scheduling, boost monitoring, and provide better overall service.

Deep Learning Ensemble for Flood Probability Analysis

Predicting flood events is complex due to uncertainties from limited gauge data, high data and computational demands of traditional physical models, and challenges in spatial and temporal scaling. This research innovatively uses only three remotely sensed and computed factors: rainfall, runoff and temperature. We also employ three deep learning models—Feedforward Neural Network (FNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM)—along with a deep neural network ensemble (DNNE) using synthetic data to predict future flood probabilities, utilizing the Savitzky–Golay filter for smoothing. Using a hydrometeorological dataset from 1993–2022 for the Nile River basin, six flood predictors were derived. The FNN and LSTM models exhibited high accuracy and stable loss, indicating minimal overfitting, while the CNN showed slight overfitting. Performance metrics revealed that FNN achieved 99.63% accuracy and 0.999886 ROC AUC, CNN had 95.42% accuracy and 0.893218 ROC AUC, and LSTM excelled with 99.82% accuracy and 0.999967 ROC AUC. The DNNE outperformed individual models in reliability and consistency. Runoff and rainfall were the most influential predictors, while temperature had minimal impact.

Leveraging Social Media Data for Forest Fires Sentiment Classification: A Data-Driven Method

Background: The rise in forest fires over the last two years, which is due to rise in dry weather conditions and human activities, have greatly impacted an area of 1.6 million hectares, leading to significant ecological, economic, and health issues, hence the need to improve disaster response strategies. Previous research determined the lack of coverage regarding public response during forest fires with conventional methods such as satellite images and sensor data. However, social media platforms provide real-time information generated by users, along with location information of disaster events. Sentiment analysis helps in understanding the public reactions and responses to natural disasters, thereby increasing awareness about forest fires. Objective: The purpose of this research is to assess the efficiency of Long Short-Term Memory (LSTM) method in classifying sentiment for social networks in regard to forest fires. This research aims to examine the effect of TF-IDF, unigram, and the FastText features on the effectiveness of the classification of sentiment. Methods: The precision, recall, and F1 score of 2, 3, and 4 determined in the LSTM models with commonly available sentiment analysis tools, such as the Vader Sentiment Analysis and SentiWordNet was used to evaluate the performance of the model. Results: With an improvement of roughly 10%, the four layers of the LSTM model generated the best performance for the evaluation of sentiments about forest fires. The LSTM model with FastText achieved F1, recall and precision scores of 0.649, 0.641, and 0.659, which exceeds lexicon-based method including SentiWordNet and Vader. Conclusion: The experimental results showed that the LSTM model outperformed lexicon-based methods when used to analyse the tweets related to forest fire. Additional research is required to integrate rule-based models and LSTM models to develop a more robust model for dynamic data. Keywords: Forest Fire, Disaster, Long Short-Term Memory, LSTM, Vader, SentiWordnet

Rancang Bangun Aplikasi Untuk Prediksi Harga Bitcoin Menggunakan Algoritma Long Short-Term Memory

Bitcoin, as the first cryptocurrency launched in 2009, has experienced rapid growth and significant volatility.The high price fluctuations of Bitcoin have attracted the attention of investors and traders worldwide.Predicting Bitcoin prices poses a challenge due to its inherent complexity and volatility. Various methodsare employed to predict Bitcoin prices, including fundamental analysis, technical analysis, and deeplearning techniques. This study explores the use of neural networks, specifically Long Short-Term Memory(LSTM), as a tool for predicting Bitcoin prices. Long Short-Term Memory is a type of recurrent neuralnetwork (RNN) designed to address the vanishing and exploding gradient problems present in traditionalRNNs and can learn long-term dependencies in data. The LSTM model used in this research comprises 50neurons, 400 epochs, a batch size of 32, and utilizes Adam optimization. Model evaluation indicates thatLSTM performs well, with a Mean Square Error (MSE) of 2688. Additionally, the model achieves a MeanAbsolute Percentage Error (MAPE) accuracy of 97.77%. Based on these predictions, Bitcoin's price isexpected to decrease over the next month, with an estimated price of approximately 66,191.00 on June 15,2024, and around 64,271.64 on July 15, 2024

Enhancing drug-target interaction predictions in context of neurodegenerative diseases using bidirectional long short-term memory in male Swiss albino mice pharmaco-EEG analysis

Background and ObjectiveEmerging diseases like Parkinson or Alzheimer's, which are not curable, endanger human mental health and are challenging to research. Drug target interactions (DTI) are pivotal in the screening of candidate drugs and focus on a small pool of drug targets. Electroencephalogram shows the responses to psychotropic medicines in the brain bioelectric activity. Synaptic activity can be analyzed by using Local Field Potential recordings obtained from micro-electrodes implanted in the brain. The aim is to evaluate the effects of drug on brain bioelectric activity and increase the drug classification accuracy. The ultimate goal is to advance our understanding of how drugs affect synaptic activity and open the door to more focused treatment for neurodegenerative diseases. MethodsIn this study, Pharmaco-EEG recordings are processed using Advanced neural network models, particularly Convolutional Neural Networks, to assess the effects of medications. The five different medicines used in this study are Ephedrine, Fluoxetine, Kratom, Morphine, and Saline. The signals observed are local field potential signals. To overcome some limits of DTI prediction, we propose Bidirectional Long Short-Term Memory (LSTM) for the categorization of intracranial EEG (i-EEG) data, departing from standard approaches. Similar EEG patterns are presumably caused by drugs that work by homologous pharmacological pathways, producing similar psychotropic effects. To improve accuracy and reduce training loss, our study introduces a bidirectional LSTM model for classification along with Bayesian optimization ResultsHigh recall, precision, and F1-Scores, particularly a 95% F1-Score for morphine, ephedrine, fluoxetine, and saline, suggest good performance in predicting these drug classes. Kratom produces a somewhat lower recall of 94%, but a high F1-Score of 97% and perfect precision of 1.00. The weighted average F1-Score, macro average, and overall accuracy are all consistently high (around 97%), indicating that the model works well throughout the spectrum of drugs. ConclusionsImproved model performance was demonstrated by using a diversified dataset with five drug categories and bidirectional LSTM boosted with Bayesian optimization for hyperparameter tuning. From earlier limited-category models, it represents a substantial advancement.

Soil and Water Assessment Tool (SWAT)-Informed Deep Learning for Streamflow Forecasting with Remote Sensing and In Situ Precipitation and Discharge Observations

In order to anticipate residual errors and improve accuracy while reducing uncertainties, this work integrates the long short-term memory (LSTM) with the Soil and Water Assessment Tool (SWAT) to create a deep learning (DL) model that is guided by physics. By forecasting the residual errors of the SWAT model, the SWAT-informed LSTM model (LSTM-SWAT) differs from typical LSTM approaches that predict the streamflow directly. Through numerical tests, the performance of the LSTM-SWAT was evaluated with both LSTM-only and SWAT-only models in the Upper Heihe River Basin. The outcomes showed that the LSTM-SWAT performed better than the other models, showing higher accuracy and a lower mean absolute error (MAE = 3.13 m3/s). Sensitivity experiments further showed how the quality of the training dataset affects the performance of the LSTM-SWAT. The results of this study demonstrate how the LSTM-SWAT may improve streamflow prediction greatly by remote sensing and in situ observations. Additionally, this study emphasizes the need for detailed consideration of specific sources of uncertainty to further improve the predictive capabilities of the hybrid model.

Enhancing Daily Crash Count Prediction using Deep Learning with Window Size Selection and Seasonality Predictor Integration

The escalating need for proactive safety measures, coupled with advancements in data collection and analytical techniques, has significantly refined the accuracy of crash count predictions, shifting from annual scales to finer daily or hourly estimates. This research places emphasis on recurrent neural networks (RNNs), specifically the long short-term memory (LSTM) model, acknowledged for effectively managing sequential data in time series predictions. Paramount considerations encompass the treatment of input data, including the decision to incorporate temporal features alongside endogenous historical target values, and the establishment of an optimal window size for data input. Despite the critical nature of these facets, exhaustive studies concurrently investigating both under controlled conditions are scarce. This research addresses this gap, assessing diverse scenarios featuring distinct temporal treatments and window sizes, and employing an LSTM model with uniform fine-tuned parameters to ensure a fair comparison. Findings indicate a significant variation in performance among models employing different window sizes and month predictor integration, under identical RNN structures and LSTM configurations. The best model outperformed the least effective by roughly 30% concerning root mean square error (RMSE) and ranking correlation between predicted and actual target crash counts in the test datasets. Unlike seasonality predictor treatment, diverse window size selections did not lead to statistically significant differences in model performance. Generally, a window size of 3 performed worst under most conditions, followed by a size of 14. Sizes of 7 and 28 performed best in nearly an equal numbers of cases.

Photometric redshift estimation for CSST survey with LSTM neural networks

ABSTRACT Accurate estimation of photometric redshifts (photo-zs) is crucial for cosmological surveys. Various methods have been developed for this purpose, such as template fitting methods and machine learning techniques, each with its own applications, advantages, and limitations. In this study, we propose a new approach that utilizes a deep learning model based on recurrent neural networks with long short-term memory (LSTM) to predict photo-z. Unlike many existing machine learning models, our method requires only flux measurements from different observed filters as input. The model can automatically learn the complex relationships between the flux data across different wavelengths, eliminating the need for manually extracted or derived input features, thereby providing precise photo-z estimates. The effectiveness of our proposed model is evaluated using simulated data from the Chinese Space Station Telescope (CSST) sourced from the Hubble Space Telescope Advanced Camera for Surveys (HST-ACS) and the COSMOS catalogue, considering anticipated instrument effects of the future CSST. Results from experiments demonstrate that our LSTM model, compared to commonly used template fitting and machine learning approaches, requires minimal input parameters and achieves high precision in photo-z estimation. For instance, when trained on the same data set and provided only with photometric fluxes as input features, the proposed LSTM model yields one-third of the outliers $f_{\mathrm{out}}$ observed with a multi-layer perceptron neural network (MLP) model, while the normalized median absolute deviation $\rm \sigma _{NMAD}$ is only two-thirds that of the MLP model. This study presents a novel approach to accurately estimate photo-zs of galaxies using photometric data from large-scale survey projects.

Federated Deep Learning Model for False Data Injection Attack Detection in Cyber Physical Power Systems

Cyber-physical power systems (CPPS) integrate information and communication technology into conventional electric power systems to facilitate bidirectional communication of information and electric power between users and power grids. Despite its benefits, the open communication environment of CPPS is vulnerable to various security attacks. This paper proposes a federated deep learning-based architecture to detect false data injection attacks (FDIAs) in CPPS. The proposed work offers a strong, decentralized alternative with the ability to boost detection accuracy while maintaining data privacy, presenting a significant opportunity for real-world applications in the smart grid. This framework combines state-of-the-art machine learning and deep learning models, which are used in both centralized and federated learning configurations, to boost the detection of false data injection attacks in cyber-physical power systems. In particular, the research uses a multi-stage detection framework that combines several models, including classic machine learning classifiers like Random Forest and ExtraTrees Classifiers, and deep learning architectures such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The results demonstrate that Bidirectional GRU and LSTM models with attention layers in a federated learning setup achieve superior performance, with accuracy approaching 99.8%. This approach enhances both detection accuracy and data privacy, offering a robust solution for FDIA detection in real-world smart grid applications.