• 대한생식의학회:학술대회논문집
• /
• 2001.11a
• /
• pp.109.2-109.2
• /
• 2001

• Journal of the Korean Mathematical Society
• /
• v.39 no.6
• /
• pp.953-961
• /
• 2002

Generalized Ricci-recurrent trans-Sasakian manifolds are studied. Among others, it is proved that a generalized Ricci-recurrent cosymplectic manifold is always recurrent Generalized Ricci-recurrent trans-Sasakian manifolds of dimension $\geq$ 5 are locally classified. It is also proved that if M is one of Sasakian, $\alpha$-Sasakian, Kenmotsu or $\beta$-Kenmotsu manifolds, which is gener-alized Ricci-recurrent with cyclic Ricci tensor and non-zero A (ξ) everywhere; then M is an Einstein manifold.

• Journal of Communications and Networks
• /
• v.5 no.1
• /
• pp.7-18
• /
• 2003

In this paper, a decision feedback recurrent neural network equalizer and a modified real time recurrent learning algorithm are proposed, and an adaptive adjusting of the learning step is also brought forward. Then, a complex case is considered. A decision feedback complex recurrent neural network equalizer and a modified complex real time recurrent learning algorithm are proposed. Moreover, weights of decision feedback recurrent neural network equalizer under burst-interference conditions are analyzed, and two anti-burst-interference algorithms to prevent equalizer from out of working are presented, which are applied to both real and complex cases. The performance of the recurrent neural network equalizer is analyzed based on numerical results.

• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.1
• /
• pp.183-193
• /
• 2024

Recurrent Neural Network technology that learns past patterns and predicts future patterns using technology for recognizing and classifying objects is being applied to various industries, economies, and languages. And research for practical use is making a lot of progress. However, research on the application of Recurrent Neural Networks for evaluating and predicting the safety of mechanical structures is insufficient. Accurate detection of external load applied to the outside is required to evaluate the safety of mechanical structures. Learning of Recurrent Neural Networks for this requires a large amount of load data. This study applied the Gated Recurrent Unit technique to examine the possibility of load learning and investigated the possibility of applying a stacked Auto Encoder as a way to secure load data. In addition, the usefulness of learning mechanical loads was analyzed with the Gated Recurrent Unit technique, and the basic setting of related functions and parameters was proposed to secure accuracy in the recognition and prediction of loads.

• Journal of Korean Society of Transportation
• /
• v.25 no.6
• /
• pp.99-109
• /
• 2007

There have been few studies on the impacts of independent recurrent and non-recurrent congestion on freeway networks. The main reason is due partly to the lack of traffic data collected during those periods of recurrent and non-recurrent congestion and partly to the difficulty of using the simulation tools effectively. This study has suggested a methodology to analyze the independent impacts of the recurrent and non-recurrent congestion on target freeway segments. The proposed methodology is based on an elaborately calibrated simulation analysis, using real traffic data obtained during the recurrent and non-recurrent congestion periods. This paper has also summarized the evaluation results from the field tests of two ITS technologies, which were developed to provide drivers with real-time traffic information under traffic congestion. As a result, their accuracy may not be guaranteed during the transition periods such as the non-recurrent congestion. In summary, this study has been focused on the importance of non-recurrent congestion compared to recurrent congestion, and the proposed methodology is expected to provide a basic foundation for prioritizing limited government investments for improving freeway network performance degraded by recurrent or non-recurrent congestion.

• Journal of Animal Reproduction and Biotechnology
• /
• v.34 no.1
• /
• pp.30-34
• /
• 2019

Milk fever is a metabolic disease with manifestation of clinical signs due to hypocalcemia, which usually occurs within 48-72 h after delivery. However, even after a successful treatment of milk fever, recurrence of milk fever may occur, and studies on recurrent milk fever are still lacking. Accordingly, the present study was conducted for the purpose of identifying the characteristics of recurrent milk fever according to farm, season, parity, and dystocia that can cause physiological changes in the mother during peri- and postpartum periods. The analysis results showed that the incidence rate of initial and recurrent milk fever according to breeding farm was 5.7%-14.1% and 3.1%-7.2%, respectively, demonstrating a positive correlation between the initial and recurrent milk fever (r = 0.613, p < 0.01). With respect to season, the incidence rate of initial and recurrent milk fever during summer was 12.3% and 7.5%, respectively, which were significantly higher than that of other seasons (p < 0.05). In addition, the recurrence rate, the ratio of recurrence relative to initial milk fever, was highest during summer with 62.7%. Regarding parity, the incidence rate of initial and recurrent milk fever in 3rd parity was 11.1% and 5.8%, respectively, which was significantly higher than in 1st and 2nd parity (p < 0.05). Furthermore, the recurrence rate in 4th parity was 64.1%, showing a pattern of increase in incidence rate with increase in parity. Finally, there were no differences in the incidence rate of initial and recurrent milk fever according to eutocia and dystocia. The findings indicated that the incidence rate of initial milk fever should be reduced to effectively prevent the recurrent milk fever, while animals with 3rd parity or higher should be expected to occur high rate of recurrent milk fever, especially during summer, and the necessary preparations should be made for intensive treatment of such individuals.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.11
• /
• pp.917-924
• /
• 2003

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. We propose decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, utilizing extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental results for two time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.390-393
• /
• 2007

In this paper, applications of self recurrent neuro-fuzzy controller to stabilization of nonlinear system are considered. The architecture of self recurrent neuro-fuzzy controller is fix layer, and the hidden layer is comprised of self recurrent architecture. Also, generalized dynamic error-backpropagation algorithm is used for the learning of the self recurrent neuro-fuzzy controller. To demonstrate the efficiency of the self recurrent neuro-fuzzy control algorithm presented in this study, a self recurrent neuro-fuzzy controller was designed and then a comparative analysis was made with LQR controller through an simulation.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.8 no.2
• /
• pp.232-234
• /
• 1997

After the first report of mitral stenosis as a cause of recurrent laryngeal nerve palsy by Ortner in 1897, many authors have described that some kinds of cardiovascular disease might contribute to the development of recurrent laryngeal nerve palsy. The estimated rate of aortic aneurysm related with recurrent laryngeal nerve palsy is about 5%. Aortic aneurysm is classified into 3 types according to the involving segment of aorta in which aneurysms develop, and the first class-aneurysm in ascending aorta and aortic arch-is known to be the only type related to recurrent laryngeal nerve palsy. Recently we experienced two cases of recurrent laryngeal nerve palsy each of which had aneurysm on aortic arch as a major contributing factor. We report these cases with brief review of the literature.

• ETRI Journal
• /
• v.39 no.6
• /
• pp.832-840
• /
• 2017

Recently, deep recurrent neural networks have achieved great success in various machine learning tasks, and have also been applied for sound event detection. The detection of temporally overlapping sound events in realistic environments is much more challenging than in monophonic detection problems. In this paper, we present an approach to improve the accuracy of polyphonic sound event detection in multichannel audio based on gated recurrent neural networks in combination with auditory spectral features. In the proposed method, human hearing perception-based spatial and spectral-domain noise-reduced harmonic features are extracted from multichannel audio and used as high-resolution spectral inputs to train gated recurrent neural networks. This provides a fast and stable convergence rate compared to long short-term memory recurrent neural networks. Our evaluation reveals that the proposed method outperforms the conventional approaches.