• Journal of Oral Medicine and Pain
• /
• v.20 no.2
• /
• pp.449-459
• /
• 1995

To investigate the relationship between recurrent aphthous ulcer and oral mucosal keratinization, exfoliative cytology in buccal mucosa, lip mucosa, tongue mucosa were performed on 25 recurrent aphthous ulcer patients and 25 controls whose age ranged from 10 to 65. Keratinization cell ratio was then measured. The results were as follows : 1. Yellow cell ratio in the control group was more than that in the patient group in buccal mucosa, lip mucosa, tongue mucosa. Red cell ratio in the control group was more than that in the patient group in lip mucosa. Blue cell ratio in the patient group was more than that in control group in all regions( p(0.01) 2. In the comparison by sex, the patient group showed no significant difference in all site but, the control group showed different results according to the site; males were more than females in yellow cell, but less than females in red cell Females were more than males in yellow cell, but less than males in red cell. 3. In the comparison by age, patient group showed no significant difference in all site, but the control group showed significantly high yellow cell ratio in buccal and tongue mucosa over the age of 50. In conclusion, there was close relationship between recurrent aphthous ulcer and decreased oral mucosal keratinization. In other words, reduced oral mucosal keratinization must be recommended for prevention of recurrent aphthous ulcer.

• Clinical and Experimental Reproductive Medicine
• /
• v.28 no.4
• /
• pp.301-305
• /
• 2001

Objective : To analyze the antithrombin II deficiency in patients with recurrent spontaneous abortion. Material and Method: The blood samples were tested by chromogenic assay to evaluate the activity of antithrombin III. Results: There was only one case of antithrombin III deficiency. This patient experienced one neonatal death after delivery and one FDIU (fetal death in utero). And also this patient showed a lupus anticoagulant and the prolongation of PTT. Conclusions: Women with recurrent miscarriage who have no obvious identified cause should consider hematologic screening. Antithrombin III deficiency could be a cause of recurrent spontaneous abortion. But the incidence is very rare in Korean patients.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.1
• /
• pp.139-147
• /
• 2003

Dynamic neural networks have been applied to diverse fields requiring temporal signal processing such as signal prediction. This paper proposes the hybrid network, composed of locally(LRNN) and globally recurrent neural networks(GRNN), to improve dynamics of multilayered recurrent networks(RNN) and then describes nonlinear adaptive prediction using the proposed network as an adaptive filter. The hybrid network consists of IIR-MLP and Elman RNN as LRNN and GRNN, respectively. The proposed network is evaluated in nonlinear signal prediction and compared with Elman RNN and IIR-MLP networks for the relative comparison of prediction performance. Experimental results show that the hybrid network performs better with respect to convergence speed and accuracy, indicating that the proposed network can be a more effective prediction model than conventional multilayered recurrent networks in nonlinear prediction for nonstationary signals.

• Communications for Statistical Applications and Methods
• /
• v.27 no.3
• /
• pp.313-326
• /
• 2020

Deep Learning is the most important key to the development of Artificial Intelligence (AI). There are several distinguishable architectures of neural networks such as MLP, CNN, and RNN. Among them, we try to understand one of the main architectures called Recurrent Neural Network (RNN) that differs from other networks in handling sequential data, including time series and texts. As one of the main tasks recently in Natural Language Processing (NLP), we consider Neural Machine Translation (NMT) using RNNs. We also summarize fundamental structures of the recurrent networks, and some topics of representing natural words to reasonable numeric vectors. We organize topics to understand estimation procedures from representing input source sequences to predict target translated sequences. In addition, we apply multiple translation models with Gated Recurrent Unites (GRUs) in Keras on English-Korean sentences that contain about 26,000 pairwise sequences in total from two different corpora, colloquialism and news. We verified some crucial factors that influence the quality of training. We found that loss decreases with more recurrent dimensions and using bidirectional RNN in the encoder when dealing with short sequences. We also computed BLEU scores which are the main measures of the translation performance, and compared them with the score from Google Translate using the same test sentences. We sum up some difficulties when training a proper translation model as well as dealing with Korean language. The use of Keras in Python for overall tasks from processing raw texts to evaluating the translation model also allows us to include some useful functions and vocabulary libraries as well.

• Journal of Power Electronics
• /
• v.16 no.2
• /
• pp.598-609
• /
• 2016

The good control performance of permanent magnet linear synchronous motor (LSM) drive systems is difficult to achieve using linear controllers because of uncertainty effects, such as fictitious forces. A backstepping control system using adaptive modified recurrent Laguerre orthogonal polynomial neural network uncertainty observer (OPNNUO) is proposed to increase the robustness of LSM drive systems. First, a field-oriented mechanism is applied to formulate a dynamic equation for an LSM drive system. Second, a backstepping approach is proposed to control the motion of the LSM drive system. With the proposed backstepping control system, the mover position of the LSM drive achieves good transient control performance and robustness. As the LSM drive system is prone to nonlinear and time-varying uncertainties, an adaptive modified recurrent Laguerre OPNNUO is proposed to estimate lumped uncertainties and thereby enhance the robustness of the LSM drive system. The on-line parameter training methodology of the modified recurrent Laguerre OPNN is based on the Lyapunov stability theorem. Furthermore, two optimal learning rates of the modified recurrent Laguerre OPNN are derived to accelerate parameter convergence. Finally, the effectiveness of the proposed control system is verified by experimental results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.2
• /
• pp.186-191
• /
• 2011

Ramp metering has been used to solve recurrent or non-recurrent congestion on many highways. However, the existing ramp metering methods cannot control non-recurrent congestion like incident and don't have any methods to solve congestion after congestion. In addition, the methods cannot solve congestion quickly because ramp metering operates independently for each ramp. In this study, we developed SARAM which is ramp metering technique with shockwave theory in order to solve the problems. In simulation from Jangsoo IC to Joongdong IC, we confirmed that speed increased by 7.32km/h and delay time reduced by 39.14sec.

• Korean Journal of Head & Neck Oncology
• /
• v.33 no.1
• /
• pp.57-59
• /
• 2017

A non-recurrent laryngeal nerve on the left side is a rare anomaly which is reported in 0.04% and it is associated with abnormal developments of the aortic arch during embryogenesis. Although the possibility is extremely low, it is important to consider the possible existence of a non-recurrent laryngeal nerve to prevent a nerve injury during thyroidectomy. We experienced a 42 year-old male with left thyroid papillary cancer who had right side aortic arch and aberrant left subclavian artery. Even though we found that this patient had a recurrent laryngeal nerve, we present this case of the right aortic arch with an aberrant left subclavian artery variation with a brief review of literature.

• Smart Media Journal
• /
• v.9 no.1
• /
• pp.45-50
• /
• 2020

With the development of deep learning, studies using artificial neural networks based on deep learning in recommendation systems are being actively conducted. Especially, the recommendation system based on RNN (Recurrent Neural Network) shows good performance because it considers the sequential characteristics of data. This study proposes a travel route recommendation system using GRU(Gated Recurrent Unit) and Session-based Parallel Mini-batch which are RNN-based algorithm. This study improved the recommendation performance through an ensemble of top1 and bpr(Bayesian personalized ranking) error functions. In addition, it was confirmed that the RNN-based recommendation system considering the sequential characteristics in the data makes a recommendation reflecting the meaning of the travel destination inherent in the travel route.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.2
• /
• pp.534-540
• /
• 1996

This paper presents an Adaptive Time-Delay Recurrent Neural Network (ATRN) for learning and recognition of temporal correlations of temporal patterns. The ATRN employs adaptive time-delays and recurrent connections, which are inspired from neurobiology. In the ATRN, the adaptive time-delays make the ATRN choose the optimal values of time-delays for the temporal location of the important information in the input parrerns, and the recurrent connections enable the network to encode and integrate temporal information of sequences which have arbitrary interval time and arbitrary length of temporal context. The ATRN described in this paper, ATNN proposed by Lin, and TDNN introduced by Waibel were simulated and applied to the chaotic time series preditcion of Mackey-Glass delay-differential equation. The simulation results show that the normalized mean square error (NMSE) of ATRN is 0.0026, while the NMSE values of ATNN and TDNN are 0.014, 0.0117, respectively, and in temporal learning, employing recurrent links in the network is more effective than putting multiple time-delays into the neurons. The best performance is attained bythe ATRN. This ATRN will be sell applicable for temporally continuous domains, such as speech recognition, moving object recognition, motor control, and time-series prediction.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.8
• /
• pp.1759-1766
• /
• 2010

The performance of noise attenuation of the adaptive active noise control algorithm is improved using the recurrent neural network. The FXLMS that has been frequently used in the active noise control is simple and has low computational load, but this method is weak to nonlinearity of the main or secondary path since it is based on the FIR linear filter method. In this paper, the recurrent neural network filter has been developed and applied to improvement of the active noise attenuation by simulation.