• ETRI Journal
• /
• v.41 no.3
• /
• pp.371-382
• /
• 2019

Machine reading comprehension is the task of understanding a given context and finding the correct response in that context. A simple recurrent unit (SRU) is a model that solves the vanishing gradient problem in a recurrent neural network (RNN) using a neural gate, such as a gated recurrent unit (GRU) and long short-term memory (LSTM); moreover, it removes the previous hidden state from the input gate to improve the speed compared to GRU and LSTM. A self-matching network, used in R-Net, can have a similar effect to coreference resolution because the self-matching network can obtain context information of a similar meaning by calculating the attention weight for its own RNN sequence. In this paper, we construct a dataset for Korean machine reading comprehension and propose an $S^2-Net$ model that adds a self-matching layer to an encoder RNN using multilayer SRU. The experimental results show that the proposed $S^2-Net$ model has performance of single 68.82% EM and 81.25% F1, and ensemble 70.81% EM, 82.48% F1 in the Korean machine reading comprehension test dataset, and has single 71.30% EM and 80.37% F1 and ensemble 73.29% EM and 81.54% F1 performance in the SQuAD dev dataset.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.120-126
• /
• 2019

Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE) has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure (AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificial intelligence (AI) network that can effectively perform pattern recognition of time series data such as the power consumption, and analyze performance of the prediction based on real household power usage data. In the performance analysis, performance comparison between the proposed GRU-based learning model and the conventional learning model of Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absolute error (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE) are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposed GRU-based learning model is greatly improved.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.165-171
• /
• 2022

Human heart rate can be used to measure exercise intensity as an important indicator. If heart rate can be predicted, exercise can be performed more efficiently by regulating the intensity of exercise in advance. In this paper, a FitRec-based prediction model is proposed for estimating running heart rate for users. Endomondo data is utilized for training the proposed prediction model. The processing algorithms for time-series data, such as LSTM(long short term memory) and GRU(gated recurrent unit), are employed to compare their performance. On the basis of simulation results, it was demonstrated that the proposed model trained with running exercise performed better than the model trained with several cardiac exercises.

• Journal of Biomedical Engineering Research
• /
• v.40 no.6
• /
• pp.242-249
• /
• 2019

Forearm electromyography (EMG) generated by wrist movements has been widely used to develop an electrical prosthetic hand, but EMG generated by finger movements has been rarely used even though 20% of amputees lose fingers. The goal of this study is to improve the classification performance of different finger movements using a deep learning algorithm, and thereby contributing to the development of a high-performance finger-based prosthetic hand. Ten participants took part in this study, and they performed seven different finger movements forty times each (thumb, index, middle, ring, little, fist and rest) during which EMG was measured from the back of the right hand using four bipolar electrodes. We extracted mean absolute value (MAV), root mean square (RMS), and mean (MEAN) from the measured EMGs for each trial as features, and a 5x5-fold cross-validation was performed to estimate the classification performance of seven different finger movements. A long short-term memory (LSTM) model was used as a classifier, and linear discriminant analysis (LDA) that is a widely used classifier in previous studies was also used for comparison. The best performance of the LSTM model (sensitivity: 91.46 ± 6.72%; specificity: 91.27 ± 4.18%; accuracy: 91.26 ± 4.09%) significantly outperformed that of LDA (sensitivity: 84.55 ± 9.61%; specificity: 84.02 ± 6.00%; accuracy: 84.00 ± 5.87%). Our result demonstrates the feasibility of a deep learning algorithm (LSTM) to improve the performance of classifying different finger movements using EMG.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.201-203
• /
• 2022

Over the past five years, the number of patients treated for diabetes has increased by 27.7% to 3.22 million, and since blood sugar is still checked through finger blood collection, continuous blood glucose measurement and blood sugar peak confirmation are difficult and painful. To solve this problem, based on blood sugar data measured for 14 days, three months of blood sugar prediction data are provided to diabetics using artificial intelligence technology.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.2
• /
• pp.57-72
• /
• 2020

The price of iron ore has continued to fluctuate with high demand and supply from many countries and companies. In this business environment, forecasting the price of iron ore has become important. This study developed the machine learning model forecasting the price of iron ore a one month after the trading events. The forecasting model used distributed lag model and deep learning models such as MLP (Multi-layer perceptron), RNN (Recurrent neural network) and LSTM (Long short-term memory). According to the results of comparing individual models through metrics, LSTM showed the lowest predictive error. Also, as a result of comparing the models using the ensemble technique, the distributed lag and LSTM ensemble model showed the lowest prediction.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.4
• /
• pp.161-167
• /
• 2019

Recently, demand forecasting techniques have been actively studied due to interest in stable power supply with surging power demand, and increase in spread of smart meters that enable real-time power measurement. In this study, we proceeded the deep learning prediction model experiments which learns actual measured power usage data of home and outputs the forecasting result. And we proceeded pre-processing with moving average method. The predicted value made by the model is evaluated with the actual measured data. Through this forecasting, it is possible to lower the power supply reserve ratio and reduce the waste of the unused power. In this paper, we conducted experiments on three types of networks: Multi Layer Perceptron (MLP), Recurrent Neural Network (RNN), and Long Short Term Memory (LSTM) and we evaluate the results of each scheme. Evaluation is conducted with following method: MSE(Mean Squared Error) method and MAE(Mean Absolute Error).

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.67-73
• /
• 2021

In this paper, in order to prevent the fall of the drone, a study was conducted to collect vibration data from the motor connected to the propeller of the drone, and to predict the abnormal vibration of the drone using recurrent neural network (RNN) and long short term memory (LSTM). In order to collect the vibration data of the drone, a vibration sensor is attached to the motor connected to the propeller of the drone to collect vibration data on normal, bar damage, rotor damage, and shaft deflection, and abnormal vibration data are collected through LSTM and RNN. The root mean square error (RMSE) value of the vibration prediction result were compared and analyzed. As a result of the comparative simulation, it was confirmed that both the predicted result through RNN and LSTM predicted the abnormal vibration pattern very accurately. However, the vibration predicted by the LSTM was found to be 15.4% lower on average than the vibration predicted by the RNN.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1770-1776
• /
• 2021

For hand expression recognition, hand pose recognition based on the static shape of the hand and hand gesture recognition based on hand movement are used together. In this paper, we proposed a hand expression recognition method that recognizes symbols based on the trajectory of a hand movement on a virtual blackboard. In order to recognize a sign drawn by hand on a virtual blackboard, not only a method of recognizing a sign from a hand movement, but also hand pose recognition for finding the start and end of data input is also required. In this paper, MediaPipe was used to recognize hand pose, and LSTM(Long Short Term Memory), a type of recurrent neural network, was used to recognize hand gesture from time series data. To verify the effectiveness of the proposed method, it was applied to the recognition of numbers written on a virtual blackboard, and a recognition rate of about 94% was obtained.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.131-131
• /
• 2021

Deep learning models, especially those based on long short-term memory (LSTM), have presented their superiority in addressing time series data issues recently. This study aims to comprehensively evaluate the performance of deep learning models that belong to the supervised learning category in streamflow prediction. Therefore, six deep learning models-standard LSTM, standard gated recurrent unit (GRU), stacked LSTM, bidirectional LSTM (BiLSTM), feed-forward neural network (FFNN), and convolutional neural network (CNN) models-were of interest in this study. The Red River system, one of the largest river basins in Vietnam, was adopted as a case study. In addition, deep learning models were designed to forecast flowrate for one- and two-day ahead at Son Tay hydrological station on the Red River using a series of observed flowrate data at seven hydrological stations on three major river branches of the Red River system-Thao River, Da River, and Lo River-as the input data for training, validation, and testing. The comparison results have indicated that the four LSTM-based models exhibit significantly better performance and maintain stability than the FFNN and CNN models. Moreover, LSTM-based models may reach impressive predictions even in the presence of upstream reservoirs and dams. In the case of the stacked LSTM and BiLSTM models, the complexity of these models is not accompanied by performance improvement because their respective performance is not higher than the two standard models (LSTM and GRU). As a result, we realized that in the context of hydrological forecasting problems, simple architectural models such as LSTM and GRU (with one hidden layer) are sufficient to produce highly reliable forecasts while minimizing computation time because of the sequential data nature.