Vanilla Recurrent Neural Networks Research Articles

Low Tropospheric Wind Forecasts in Aviation: The Potential of Deep Learning for Terminal Aerodrome Forecast Bulletins

In aviation, accurate wind prediction is crucial, especially during takeoff and landing at complex sites like Gran Canaria Airport. This study evaluated five Deep Learning models: Long Short-Term Memory (LSTM), Vanilla Recurrent Neural Network (vRNN), One-Dimensional Convolutional Neural Network (1dCNN), Convolutional Neural Network Long Short-Term Memory (CNN-LSTM), and Gated Recurrent Unit (GRU) for forecasting wind speed and direction. The LSTM model demonstrated the highest precision, particularly for extended forecasting periods, achieving a mean absolute error (MAE) of 1.23 m/s and a circular MAE (cMAE) of 15.80° for wind speed and direction, respectively, aligning with World Meteorological Organization standards for Terminal Aerodrome Forecasts (TAF). While the GRU and CNN-LSTM also showed promising results, and the 1dCNN excelled in wind direction forecasting over shorter intervals, the vRNN lagged in performance. Additionally, the autoregressive integrated moving average model underperformed relative to the DL models, underscoring the potential of DL, particularly LSTM, in enhancing TAF accuracy at airports with intricate wind patterns. This study not only confirms the superiority of DL over traditional methods but also highlights the promise of integrating artificial intelligence into TAF automation.

A robust approach for aspect-based sentiment analysis using deep learning and domain ontologies

PurposeThe purpose of this research study is to improve sentiment analysis (SA) at the aspect level, which is accomplished through two independent goals of aspect term and opinion extraction and subsequent sentiment classification.Design/methodology/approachThe proposed architecture uses neighborhood and dependency tree-based relations for target opinion extraction, a domain–ontology-based knowledge management system for aspect term extraction, and deep learning techniques for classification.FindingsThe authors use different deep learning architectures to test the proposed approach of both review and aspect levels. It is reported that Vanilla recurrent neural network has an accuracy of 83.22%, long short-term memory (LSTM) is 89.87% accurate, Bi-LSTM is 91.57% accurate, gated recurrent unit is 65.57% accurate and convolutional neural network is 82.33% accurate. For the aspect level analysis, ρaspect comes out to be 0.712 and Δ2aspect is 0.384, indicating a marked improvement over previously reported results.Originality/valueThis study suggests a novel method for aspect-based SA that makes use of deep learning and domain ontologies. The use of domain ontologies allows for enhanced aspect identification, and the use of deep learning algorithms enhances the accuracy of the SA task.

Utilizing sequential modeling in collaborative method for flood forecasting

Flood forecasting has been a major challenge in hydrology for decades. A variety of approaches have been developed, including numerical models and data-driven models, such as Deep Learning (DL), for disaster early-warning and monitoring systems. This study proposed two combination models, exploring different feature separation techniques to improve the 12-hour multi-step discharge prediction. Specifically, two different approaches were implemented, namely the separation of wet and dry periods (Combination Model No. 1; CM-1) and the separation of flood and non-flood events (Combination Model No. 2; CM-2) for comparison with the baseline model trained on an entire dataset. Multivariate time series discharge data were used, measured at upstream and downstream stations within the same river basins. Experiments were conducted in three case study areas in Thailand: the Lower Loei River Basin, the Upper Nan River Basin, and the Upper Chi River Basin.In the proposed model combination approach, three deep-learning models were investigated: vanilla recurrent neural networks (RNNs) and their variants, long short-term memory networks (LSTMs), and gated recurrent units (GRUs). In general, CM-2 exhibited superior performance compared to the other models in most of the experimented neural networks, while the CM-1 approach outperformed the baseline model. This emphasized the advantages of our proposed feature separation framework, particularly in improving model performance, with respect to flood events. However, some variations were observed in the Upper Chi River Basin because the more complex models, such as LSTMs and GRUs, possibly overfitted the training set with less fluctuating data. Our study highlighted the significant contribution of combining two models through feature separation, tailored to specific periods, thereby enhancing the accuracy of flood predictions.

Recurrent Neural Networks for Image Captioning: A Case Study with LSTM

This research investigates the viability of Long Short-Term Memory (LSTM) systems, a subtype of Recurrent Neural Networks (RNNs), for picture captioning. Leveraging the MS COCO dataset, the study compares the execution of LSTM-based RNNs with Vanilla RNN, Gated Recurrent Unit (GRU), consideration components, and transformer-based models. Experimental comes about to illustrate that the LSTM-based RNN shows competitive execution, accomplishing a BLEU-4 score of 0.72, a METEOR score of 0.68, and a CIDEr score of 2.1. The comparative investigation uncovers its prevalence over Vanilla RNN and GRU, highlighting its capability to capture long-range conditions inside successive picture information. Moreover, the study investigates the effect of consideration instruments and transformer designs, exhibiting their potential improvements in the context-aware caption era. The transformer-based show outflanks all other models, accomplishing a BLEU-4 score of 0.78, a METEOR score of 0.72, and a CIDEr score of 2.5. The findings give important bits of knowledge toward the creating scene of picture captioning strategy, which makes LSTM-based RNNs solid and productive approaches for capturing worldly groupings in visual substance. In achieving these, the study provides a framework for future developments in hybrid models and manufacturing processes that push boundaries of smart image perception and understanding

A Systematic Evaluation of Recurrent Neural Network Models for Edge Intelligence and Human Activity Recognition Applications

The Recurrent Neural Networks (RNNs) are an essential class of supervised learning algorithms. Complex tasks like speech recognition, machine translation, sentiment classification, weather prediction, etc., are now performed by well-trained RNNs. Local or cloud-based GPU machines are used to train them. However, inference is now shifting to miniature, mobile, IoT devices and even micro-controllers. Due to their colossal memory and computing requirements, mapping RNNs directly onto resource-constrained platforms is arcane and challenging. The efficacy of edge-intelligent RNNs (EI-RNNs) must satisfy both performance and memory-fitting requirements at the same time without compromising one for the other. This study’s aim was to provide an empirical evaluation and optimization of historic as well as recent RNN architectures for high-performance and low-memory footprint goals. We focused on Human Activity Recognition (HAR) tasks based on wearable sensor data for embedded healthcare applications. We evaluated and optimized six different recurrent units, namely Vanilla RNNs, Long Short-Term Memory (LSTM) units, Gated Recurrent Units (GRUs), Fast Gated Recurrent Neural Networks (FGRNNs), Fast Recurrent Neural Networks (FRNNs), and Unitary Gated Recurrent Neural Networks (UGRNNs) on eight publicly available time-series HAR datasets. We used the hold-out and cross-validation protocols for training the RNNs. We used low-rank parameterization, iterative hard thresholding, and spare retraining compression for RNNs. We found that efficient training (i.e., dataset handling and preprocessing procedures, hyperparameter tuning, and so on, and suitable compression methods (like low-rank parameterization and iterative pruning) are critical in optimizing RNNs for performance and memory efficiency. We implemented the inference of the optimized models on Raspberry Pi.

Variations of Training Process in Vanilla Recurrent Neural Network Framework

Recurrent neural networks (RNNs) are capable of learning features and long term dependencies from sequential and time series data and show outstanding performance in sequential modeling tasks. However, training process in RNNs is troubled by issues in learning processes such as slow inference, vanishing gradients and difficulties in capturing long term dependencies. In this paper, we introduce a new learning technique to update the weight set as we change the input sequence which is shifted by certain amount of time in training process, instead of using a traditional way to calculate one set of the weights and bias in training time series with sequences shifted by certain amount of time series. We also consider an algorithm for an evaluation process. In the traditional way, the evaluation process is executed by using final weights and biases calculated in the training process. Instead, during the testing process, the weights and biases are iteratively updated in each sequence as done in the training process. Several numerical experiments demonstrate the efficiency of the proposed techniques.

Tsunami tide prediction in shallow water using recurrent neural networks: model implementation in the Indonesia Tsunami Early Warning System

Near-field tides prediction for tsunami detection in the coastal area is a significant problem of the cable-based tsunami meter system in north Sipora, Indonesia. The problem is caused by its shallow water condition and the unavailability of an applicable model or research for tsunami detection in this area. The problem foundation of shallow water area is its ambient noise level-dependent property that requires preprocessing to improve its feature representation. Moreover, because this shallow water is close to the land area, we must consider a model that can accommodate low prediction time for a Tsunami Early Warning System. Therefore, we propose a recurrent neural network (RNN) model because of its reliable performance for time series forecasting. Our report evaluates variants of the RNN model (the vanilla RNN, LSTM and GRU models) in tides prediction and z-score analysis for tsunami identification. The GRU model overwhelms the other two variants in error scores and time processed (training and prediction). It can achieve median error score distribution of 7.8×10-5\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$7.8 \ imes 10^{-5}$$\\end{document} on the L1000-P250 configuration with time prediction under 0.1 s. This lower-time prediction is necessary to ensure the early warning system is going well. Moreover, the GRU model can identify all synthetic tsunami tide spikes (compared with the ground truth result) from magnitude 7.2–8.2 by applying a z-score on the GRU’s prediction.

A Deep Multi-Task Learning Approach for Bioelectrical Signal Analysis

Deep learning is a promising technique for bioelectrical signal analysis, as it can automatically discover hidden features from raw data without substantial domain knowledge. However, training a deep neural network requires a vast amount of labeled samples. Additionally, a well-trained model may be sensitive to the study object, and its performance may deteriorate sharply when transferred to other study objects. We propose a deep multi-task learning approach for bioelectrical signal analysis to address these issues. Explicitly, we define two distinct scenarios, the consistent source-target scenario and the inconsistent source-target scenario based on the motivation and purpose of the tasks. For each scenario, we present methods to decompose the original task and dataset into multiple subtasks and sub-datasets. Correspondingly, we design the generic deep parameter-sharing neural networks to solve the multi-task learning problem and illustrate the details of implementation with one-dimension convolutional neural networks (1D CNN), vanilla recurrent neural networks (RNN), recurrent neural networks with long short-term memory units (LSTM), and recurrent neural networks with gated recurrent units (GRU). In these two scenarios, we conducted extensive experiments on four electrocardiogram (ECG) databases. The results demonstrate the benefits of our approach, showing that our proposed method can improve the accuracy of ECG data analysis (up to 5.2%) in the MIT-BIH arrhythmia database.

Theoretical Foundation and Design Guideline for Reservoir Computing-Based MIMO-OFDM Symbol Detection

In this paper, we derive a theoretical upper bound on the generalization error of reservoir computing (RC), a special category of recurrent neural networks (RNNs). The specific RC implementation considered in this paper is the echo state network (ESN), and an upper bound on its generalization error is derived via the empirical Rademacher complexity (ERC) approach. While recent work in deriving risk bounds for RC frameworks makes use of a non-standard ERC measure and a direct application of its definition, our work uses the standard ERC measure and tools allowing fair comparison with conventional RNNs. The derived result shows that the generalization error bound obtained for ESNs is tighter than the existing bound for vanilla RNNs, suggesting easier generalization for ESNs. With the ESN applied to symbol detection in MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) systems, we show how the derived generalization error bound can guide underlying system design. Specifically, the derived bound together with the empirically characterized training loss is utilized to identify the optimum reservoir size in neurons for the ESN-based symbol detector. Finally, we corroborate our theoretical findings with results from simulations that employ 3GPP standards-compliant wireless channels, signifying the practical relevance of our work.

MINT: Detecting Fraudulent Behaviors from Time-Series Relational Data

The e-commerce platforms, such as Shopee, have accumulated a huge volume of time-series relational data, which contains useful information on differentiating fraud users from benign users. Existing fraud behavior detection approaches typically model the time-series data with a vanilla Recurrent Neural Network (RNN) or combine the whole sequence as a single intention without considering the temporal behavioral patterns, row-level interactions, and different view intentions. In this paper, we present MINT, a M ultiview row- IN teractive T ime-aware framework to detect fraudulent behaviors from time-series structured data. The key idea of MINT is to build a time-aware behavior graph for each user's time-series relational data with each row represented as an action node. We utilize the user's temporal information to construct three different graph convolutional matrices for hierarchically learning the user's intentions from different views, that is, short-term, medium-term, and long-term intentions. To capture more meaningful row-level interactions and alleviate the over-smoothing issue in a vanilla time-aware behavior graph, we propose a novel gated neighbor interaction mechanism to calibrate the aggregated information by each action node. Since the receptive fields of the three graph convolutional layers are designed to grow nearly exponentially, our MINT requires many fewer layers than traditional deep graph neural networks (GNNs) to capture multi-hop neighboring information, and avoids recurrent feedforward propagation, thus leading to higher training efficiency and scalability. Our extensive experiments on the large-scale e-commerce datasets from Shopee with up to 4.6 billion records and a public dataset from Amazon show that MINT achieves superior performance over 10 state-of-the-art models and provides better interpretability and scalability.

Automated model discovery for muscle using constitutive recurrent neural networks

The stiffness of soft biological tissues not only depends on the applied deformation, but also on the deformation rate. To model this type of behavior, traditional approaches select a specific time-dependent constitutive model and fit its parameters to experimental data. Instead, a new trend now suggests a machine-learning based approach that simultaneously discovers both the best model and best parameters to explain given data. Recent studies have shown that feed-forward constitutive neural networks can robustly discover constitutive models and parameters for hyperelastic materials. However, feed-forward architectures fail to capture the history dependence of viscoelastic soft tissues. Here we combine a feed-forward constitutive neural network for the hyperelastic response and a recurrent neural network for the viscous response inspired by the theory of quasi-linear viscoelasticity. Our novel rheologically-informed network architecture discovers the time-independent initial stress using the feed-forward network and the time-dependent relaxation using the recurrent network. We train and test our combined network using unconfined compression relaxation experiments of passive skeletal muscle and compare our discovered model to a neo Hookean standard linear solid, to an advanced mechanics-based model, and to a vanilla recurrent neural network with no mechanics knowledge. We demonstrate that, for limited experimental data, our new constitutive recurrent neural network discovers models and parameters that satisfy basic physical principles and generalize well to unseen data. We discover a Mooney–Rivlin type two-term initial stored energy function that is linear in the first invariant I1 and quadratic in the second invariant I2 with stiffness parameters of 0.60 kPa and 0.55 kPa. We also discover a Prony-series type relaxation function with time constants of 0.362s, 2.54s, and 52.0s with coefficients of 0.89, 0.05, and 0.03. Our newly discovered model outperforms both the neo Hookean standard linear solid and the vanilla recurrent neural network in terms of prediction accuracy on unseen data. Our results suggest that constitutive recurrent neural networks can autonomously discover both model and parameters that best explain experimental data of soft viscoelastic tissues. Our source code, data, and examples are available at https://github.com/LivingMatterLab.

Remaining useful lifetime estimation for discrete power electronic devices using physics-informed neural network

Estimation of Remaining Useful Lifetime (RUL) of discrete power electronics is important to enable predictive maintenance and ensure system safety. Conventional data-driven approaches using neural networks have been applied to address this challenge. However, due to ignoring the physical properties of the target RUL function, neural networks can result in unreasonable RUL estimates such as going upwards and wrong endings. In the paper, we apply the fundamental principle of Physics-Informed Neural Network (PINN) to enhance Recurrent Neural Network (RNN) based RUL estimation methods. Through formulating proper constraints into the loss function of neural networks, we demonstrate in our experiments with the NASA IGBT dataset that PINN can make the neural networks trained more realistically and thus achieve performance improvements in estimation error and coefficient of determination. Compared to the baseline vanilla RNN, our physics-informed RNN can improve Mean Squared Error (MSE) of out-of-sample estimation on average by 24.7% in training and by 51.3% in testing; Compared to the baseline Long Short Term Memory (LSTM, a variant of RNN), our physics-informed LSTM can improve MSE of out-of-sample estimation on average by 15.3% in training and 13.9% in testing.

A comparative study of retrieval-based and generative-based chatbots using Deep Learning and Machine Learning

Increased screen time may cause significant health impacts, including harmful effects on mental health. Studies on the association between technological obsessions and their influence on health have been conducted using Deep Learning (DL) and Machine Learning (ML) techniques. The deployment of chatbots in different industries has been proven as a game-changer. We study conversational Artificial Intelligence (AI) systems enabling operators to conduct conversations with machines that resemble those with humans. We design and develop two retrieval-based and generative-based chatbots, each with six designs. Among the retrieval-based chatbots, Vanilla Recurrent Neural Network (RNN) has an accuracy of 83.22%, Long Short Term Memory (LSTM) is 89.87% accurate, Bidirectional LSTM (Bi-LSTM) is 91.57% accurate, Gated Recurrent Unit (GRU) is 65.57% accurate, and Convolution Neural Network (CNN) is 82.33% accurate. In comparison, generative-based chatbots have encoder–decoder designs that are 94.45% accurate. The most significant distinction is that while generative-based chatbots can generate new text, retrieval-based chatbots are restricted to responding to inputs that match the best of the outputs they already know.

Accurate online training of dynamical spiking neural networks through Forward Propagation Through Time

With recent advances in learning algorithms, recurrent networks of spiking neurons are achieving performance that is competitive with vanilla recurrent neural networks. However, these algorithms are limited to small networks of simple spiking neurons and modest-length temporal sequences, as they impose high memory requirements, have difficulty training complex neuron models and are incompatible with online learning. Here, we show how the recently developed Forward-Propagation Through Time (FPTT) learning combined with novel liquid time-constant spiking neurons resolves these limitations. Applying FPTT to networks of such complex spiking neurons, we demonstrate online learning of exceedingly long sequences while outperforming current online methods and approaching or outperforming offline methods on temporal classification tasks. The efficiency and robustness of FPTT enable us to directly train a deep and performant spiking neural network for joint object localization and recognition, demonstrating the ability to train large-scale dynamic and complex spiking neural network architectures. Memory efficient online training of recurrent spiking neural networks without compromising accuracy is an open challenge in neuromorphic computing. Yin and colleagues demonstrate that training a recurrent neural network consisting of so-called liquid time-constant spiking neurons using an algorithm called Forward-Propagation Through Time allows for online learning and state-of-the-art performance at a reduced computational cost compared with existing approaches.

Full waveform inversion as training a neural network

Full waveform inversion (FWI) is usually solved as a nonlinear least-squares problem to minimize the discrepancy between the recorded signal and the synthetic data by some gradient-based optimization methods. In this paper, we investigate acoustic FWI as training a neural network. We recast the time-domain difference scheme into a forward propagation process of vanilla recurrent neural network (RNN) and then find that the parameters of RNN coincide with the physical parameters in the wave equation. As a result, the FWI problem is resolved as training such a kind of neural network. Some stochastic optimization methods such as the Adam optimizer are applied to update the model parameters. We demonstrate our method numerically with three typical models including the benchmark Marmousi model. The numerical results show that the fast convergence and good velocity inversion result can be achieved.

Robust recurrent neural networks for time series forecasting

Recurrent neural networks (RNNs) are widely utilized in time series forecasting tasks. In practical applications, there are noises in real-life time series data. A model’s generalization capacity will be diminished by model uncertainty with regard to input noises. However, the robustness of RNNs with respect to input noises has not been well studied yet. The localized stochastic sensitivity (LSS), which measures output disturbances with respect to input perturbations of learning models, has been successfully applied to improve the robustness of different neural networks on tabular and image data. But its effectiveness on time series data has not been explored. Therefore, we extend the idea of LSS and apply it to the robust RNNs training for time series forecasting problems. With the minimization of LSS, output sensitivities of RNNs with respect to small perturbations are reduced. So, the proposed robust RNNs will not be affected by slight input noises. We have used the LSTM as an example for theoretical analysis and analyzed the effectiveness of LSS on several RNN variants including the vanilla RNN, the gated recurrent unit (GRU), the long-short-term memory (LSTM), and the bi-directional long short-term memory (Bi-LSTM) in empirical studies. Experimental results confirm the efficiency of applying LSS to enhance the robustness of RNNs for time series data. For example, the robust RNNs outperform their counterpart by 53.26% and 49.45% on average in terms of Root Mean Squared Error and R-Square on five datasets.

Air Quality Prediction Using the Fractional Gradient-Based Recurrent Neural Network.

In this study, the air quality index (AQI) of Indian cities of different tiers is predicted by using the vanilla recurrent neural network (RNN). AQI is used to measure the air quality of any region which is calculated on the basis of the concentration of ground-level ozone, particle pollution, carbon monoxide, and sulphur dioxide in air. Thus, the present air quality of an area is dependent on current weather conditions, vehicle traffic in that area, or anything that increases air pollution. Also, the current air quality is dependent on the climate conditions and industrialization in that area. Thus, the AQI is history-dependent. To capture this dependency, the memory property of fractional derivatives is exploited in this algorithm and the fractional gradient descent algorithm involving Caputo's derivative has been used in the backpropagation algorithm for training of the RNN. Due to the availability of a large amount of data and high computation support, deep neural networks are capable of giving state-of-the-art results in the time series prediction. But, in this study, the basic vanilla RNN has been chosen to check the effectiveness of fractional derivatives. The AQI and gases affecting AQI prediction results for different cities show that the proposed algorithm leads to higher accuracy. It has been observed that the results of the vanilla RNN with fractional derivatives are comparable to long short-term memory (LSTM).

ML-Assisted Equalization for 50-Gb/s/λ O-Band CWDM Transmission Over 100-km SMF

We propose and demonstrate a bidirectional Vanilla recurrent neural network (Vanilla-RNN) based equalization scheme for O-band coarse wavelength division multiplexed (CWDM) transmission. Based on a 4 × 50-Gb/s intensity modulation and direct detection (IM/DD) system, we demonstrate the significantly better bit error rate (BER) performance of the Vanilla-RNN scheme over the conventional decision feedback equalizer (DFE) for both Nyquist on-off keying (OOK) and Nyquist 4-ary pulse amplitude modulation (PAM4) formats. It is shown that the Vanilla-RNN equalizer is capable of compensating for both linear and nonlinear impairments induced by the transceiver and the single-mode fiber (SMF). As a result, up to 100-km and 75-km SMF transmission can be achieved for OOK and PAM4 transmission, respectively. Furthermore, through the comparison with other equalization schemes, including the linear equalizer, 3 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rd</sup> -order Volterra equalizer, and Volterra+DFE, it is demonstrated that the Vanilla-RNN equalizer achieves the best BER performance. In the meantime, it also exhibits lower implementation complexity when compared to Volterra-based schemes. Our results show that the Vanilla-RNN scheme is a viable solution for realizing simple and effective equalization. This work serves as an exploration and offers useful insights for future implementations of reach-extended O-band CWDM IM/DD systems.

Memory augmented recurrent neural networks for de-novo drug design.

A recurrent neural network (RNN) is a machine learning model that learns the relationship between elements of an input series, in addition to inferring a relationship between the data input to the model and target output. Memory augmentation allows the RNN to learn the interrelationships between elements of the input over a protracted length of the input series. Inspired by the success of stack augmented RNN (StackRNN) to generate strings for various applications, we present two memory augmented RNN-based architectures: the Neural Turing Machine (NTM) and the Differentiable Neural Computer (DNC) for the de-novo generation of small molecules. We trained a character-level convolutional neural network (CNN) to predict the properties of a generated string and compute a reward or loss in a deep reinforcement learning setup to bias the Generator to produce molecules with the desired property. Further, we compare the performance of these architectures to gain insight to their relative merits in terms of the validity and novelty of the generated molecules and the degree of property bias towards the computational generation of de-novo drugs. We also compare the performance of these architectures with simpler recurrent neural networks (Vanilla RNN, LSTM, and GRU) without an external memory component to explore the impact of augmented memory in the task of de-novo generation of small molecules.

딥러닝 기반 LSTM 모형에 기초한 자기상관 공정의 모니터링 절차

자기상관 공정을 모니터링하는 절차에 대해서는 많은 연구가 진행되어 왔는데, 그중에서 가장 많이 사용하는 방법은 관측값을 예측하여 잔차를 계산하고 그 잔차 데이터를 모니터링하는 것이다. 즉, 잔차를 이용함으로써 자기상관 문제를 해결하는 것이다. 이 논문에서는 예측하는 방법으로 딥러닝에 기반한 LSTM (long short-term memory) 모형을 사용하여 자기상관 공정을 모니터링하는 절차를 제안하고, 제안된 절차와 기본적인 RNN (vanilla recurrent neural network) 모형을 사용하는 절차 및 시계열 모형에 적합시키는 절차의 성능을 모의실험을 수행하여 비교하였다. 딥러닝 알고리즘을 기반으로 한 LSTM과 RNN 모형에 기초한 절차가 전반적으로 좋은 성능을 나타냈으며, 시계열 모형에 적합시키는 절차에 비해 정확한 모형 적합의 과정이 필요 없다는 측면에서 효율적인 절차라고 판단된다.