논문 : 장단기 시장이자율의 장기기억에 관한 연구

이창호 ( Chang Ho Lee ),김종선 ( Chong Sun Kim ) 명지대학교 금융지식연구소 2012 금융지식연구 Vol.10 No.3

본 연구의 ARFIMA(p,d,q) 모형을 이용한 실증분석 결과는 다음과 같다. 첫째로 자기상관계수의 추정결과, 단기금리인 콜금리의 장기기억현상은 존재하였으나, 장 기금리인 국고채3년 및 회사채3년 금리에는 단위근 현상이 존재함으로써 과거의 충격이 지속되는 것으로 분석되었다. 둘째로 단기금리의 경우, 콜금리의 장기기억현상이 존재하는 것으로 분석되었으며, 글로벌금융위기가 포함된 일별콜금리의 제2기의 장기기억현상이 글로벌금융위기 이전의 제1기의 장기기억현상보다 상대적으로 강하게 나타났다. 셋째로 장기금리의 경우, 국고채3년 및 회사채3년 금리에는 대체로 단위근현상이 나타나서 장기기억현상이 존재하지 않았다. The main object of this study is to analyse the long memory effect of long-term & short-term interest rates in the point of market efficiency. Accordingly, this study focuses on analysing the ARFIMA model which detect the long memory effect of long-term(3 years-corporate bond rate & 3 years-government bond rate) and short-term(call rate) interest rates. Some empirical results are summarized as follows ; Firstly, contrary to the case of long-term interest rate, a long memory of the short-term interest rate was found in terms of autocorrelation coefficients & partial autocorrelation coefficients. Secondly, the long memory of short-term interest rate was found, and the long memory effect of the first period(2000.1∼2008.6) was stronger than that of the second period(2008.7∼2012.10). Thirdly, the long memory of long-term interest rate was not found, and the results showed the unit root process. Therefore, the market efficiency hypothesis was rejected by these results in case of short-term interest & long-term interest in Korean financial market.

Improvement of Attention, Long-term and Short-term Memory by Brain Factor-7™ (BF-7™)

Hong Junkee,노유훈,이지원,Whang Wan Kyunn,Zheng Yulong,원무호,Kang Il-Jun,김성수 건강기능식품미래포럼 2022 건강기능식품미래포럼 학술지 Vol.2 No.1

Brain Factor-7™ (BF-7™) is a mixture of silk peptides obtained from hydrolyzed fibroin of the cocoon shell of Bombyx mori, which was developed by Famenity Co., LTD and approved as a health functional food by The Korea Ministry of Food and Drug safety. Several previous clinical studies showed that BF-7™ enhanced the learning and cognitive function in various age groups. In the present study, a clinical study was performed to assess whether BF-7™ enhances short-term, long-term memory and attention on 28 college students who were given 400 mg of BF-7™ orally once a day for 4 weeks. The memory was evaluated by Korean version of Memory Assessment Scales (K-MAS) and attention was by encephalogram of P300 wave. The results were as follows. Short-term verbal memory, short-term visual memory, long-term verbal memory and long-term word memory improved by 47.2, 42.2, 54.8 and 22.3%, respectively. The encephalogram of P300 wave showed that the attention level was enhanced significantly but with less stress. In the in vitro studies on SHSY-5Y cells (a neuronal cell line), BF-7™ were shown to prevent the toxic effects of Aβ1-42 on these cells such as the decrease of cell viability, apoptosis, decrease of mitochondrial membrane potential and generation of reactive oxygen species. These results suggest that BF-7™ enhances both short- and long-term memory as well as attention level and prevents harmful actions of Aβ that affects function and health of the brain.

Long Short-Term Memory를 이용한 부산항 조위 예측

김해림,전용호,박재형,윤한삼 해양환경안전학회 2022 해양환경안전학회지 Vol.28 No.4

This study developed a Recurrent Neural Network model implemented through Long Short-Term Memory (LSTM) that generates long-term tidal level data at Busan Port using tide observation data. The tide levels in Busan Port were predicted by the Korea Hydrographic and Oceanographic Administration (KHOA) using the tide data observed at Busan New Port and Tongyeong as model input data. The model was trained for one month in January 2019, and subsequently, the accuracy was calculated for one year from February 2019 to January 2020. The constructed model showed the highest performance with a correlation coefficient of 0.997 and a root mean squared error of 2.69 cm when the tide time series of Busan New Port and Tongyeong were inputted together. The study’s finding reveal that long-term tidal level data prediction of an arbitrary port is possible using the deep learning recurrent neural network model. 본 연구는 조위 관측자료를 이용하여 부산항에서의 장기 조위 자료를 생성하는 Long Short-Term Memory (LSTM)으로 구현된 순환신경망 모델을 개발하였다. 국립해양조사원의 부산 신항과 통영에서 관측된 조위 자료를 모델 입력 자료로 사용하여 부산항의 조위를 예측하였다. 모델에 대하여 2019년 1월 한 달의 학습을 수행하였으며, 이후 2019년 2월에서 2020년 1월까지 1년에 대하여 정확도를 계산하였다. 구축된 모델은 부산 신항과 통영의 조위 시계열을 함께 입력한 경우에 상관계수 0.997 및 평균 제곱근 오차 2.69m로 가장 성능이 높았다. 본 연구 결과를 바탕으로 딥러닝 순환신경망 모델을 이용하여 임의 항만의 장기 조위 자료 예측이 가능함을 알 수 있었다.

기억처리과정의 이해

조수진 한국청각언어재활학회 2012 Audiology and Speech Research Vol.8 No.1

Memory is our ability to encode, store and retain in the human brain. Generally, there are three stages in human memory processing, which are sensory memory, short-term memory, and long-term memory. Recently, researchers tend to use the new concept of “working memory” for replacing or including the old concept of short-term memory. “Working memory” emphasizes on the manipulation of information instead of not using passive maintenance. Therefore, it is critical for cognitive information, speech perception and language learning. Based on numerous research, training of auditory working memory is able to improve some selective areas of cognitive and speech-language development. Taken together, it is needed to develop training program of auditory working memory in aural rehabilitation for hearing impaired listeners.

기억의 신경생물학적 기전

정영인,이영민,문은수 대한노인정신의학회 2016 노인정신의학 Vol.20 No.1

Memory is one of the most important mental mechanisms which is crucial for us to adapt to environmental surroundings and to maintain our identity. The neurobiological mechanisms for memory are based upon the synaptic plasticity that involve both functional and structural changes at the synapses in the neural circuits participating in learning and memory. Memory is not a single process but has two forms of short-term and long-term memory that are two independent but overlapping processes that blend into one another. The short-term memory depends upon the functional change of synaptic strength but the long-term memory requires anatomic changes of synapses in the neural circuit. Memory storage seems to use elements of a common genetic switch, involving cyclic adenosine monophospate (cAMP)-dependent protein kinase, mitogen activated protein kinase, and cAMP response elementbinding protein, to convert short-term memory into long-term memory.

딥러닝 기반 LSTM 모형을 이용한 항적 추적성능 향상에 관한 연구

황진하(Jin-Ha Hwang),이종민(Jong-Min Lee) 한국정보통신학회 2021 한국정보통신학회 종합학술대회 논문집 Vol.25 No.1

항적추적 기술에 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용하는 연구로서 기존의 항적추적기술의 경우, 항공기의 등속, 등가속, 급기동, 선회(3D) 비행 등 비행 특성에 따른 칼만 필터 기반의 LMIPDA를 활용한 실시간 항적 추적 시 등속, 등가속, 급기동, 선회(3D) 비행 가중치가 자동으로 변경된다. 이러한 과정에서 등속 비행 중 급기동 비행과 같이 비행 특성이 변경될 때, 항적 손실 및 항적 추적 성능이 하락하여 비행 특성 가중치 변경성능을 향상시킬 필요성이 있다. 본 연구는 레이더의 오차 모델이 적용된 시뮬레이터의 Plot과 표적을 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용하여 학습시키고, 칼만 필터를 활용한 항적추적 결과와 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용한 항적추적결과를 비교함으로써 미리 비행 특성의 변경과정을 예측하여 등속, 등가속, 급기동, 선회(3D) 비행 가중치변경을 신속하게 함으로써 항적추적성능을 향상하기 위한 연구이다. This study applies a deep learning-based long short-term memory(LSTM) model to track tracking technology. In the case of existing track tracking technology, the weight of constant velocity, constant acceleration, stiff turn, and circular(3D) flight is automatically changed when tracking track in real time using LMIPDA based on Kalman filter according to flight characteristics of an aircraft such as constant velocity, constant acceleration, stiff turn, and circular(3D) flight. In this process, it is necessary to improve performance of changing flight characteristic weight, because changing flight characteristics such as stiff turn flight during constant velocity flight could incur the loss of track and decreasing of the tracking performance. This study is for improving track tracking performance by predicting the change of flight characteristics in advance and changing flight characteristic weigh rapidly. To get this result, this study makes deep learning-based Long Short-Term Memory(LSTM) model study the plot and target of simulator applied with radar error model, and compares the flight tracking results of using Kalman filter with those of deep learning-based Long Short-Term memory(LSTM) model.

한국어 어휘 학습에서 기억 강화 전략이 어휘 기억에 미치는 영향

남상은(Nam Sang-eun),김영주(Kim Young-joo) 국어국문학회 2011 국어국문학 Vol.- No.157

This study aims at proving the effect of using memorial strategy on learners' vocabulary memory, presuming the need of a strategy in vocabulary learning. The main results of this research were as follows; Firstly, post immediate and delayed tests were conducted in order to measure short term and long term memory of experimental group with students who used vocabulary learning strategies and the control group with students who did not use special vocabulary learning strategy. The results of post immediate test were high in the experimental group in the first, second, and third experiments, but the differences of average between two groups were not significant. However, post delayed tests showed that the average of the experimental group was higher than that of control group and the differences of the average were significant. This confirmed that use of vocabulary learning strategy was more effective in learners' long term memory. Secondly, use of the strategy of connecting associational words did not have significant effect in short term memory, but it was very positive in long term memory. Thirdly, the strategy of making sentences was not effective in short or long term memory. Consequently, this research provided suggestions that using vocabulary learning strategy on learning vocabulary was effective, thus highly recommended in Korean language education.

기계 장비의 장단기 운용을 고려한 Long Short-Term Memory 기법에 의한 하중 인식

강정호(Jung Ho Kang) 한국기계가공학회 2024 한국기계가공학회지 Vol.23 No.2

Artificial Neural Network has been developed to enable the intelligence of mechanical equipment and devices. However, research on the intelligence of mechanical structures is rare. This study examined the possibility of learning structural loads, which is an important factor in the design and analysis of machines, using long short-term memory (LSTM), which is a Recurrent Neural Network. Because the machine structure operates for a long time, the investigation of the possibility of load learning using Long Short-Term Memory with a memory function for the short and long terms can have important implications for recognizing and predicting accidents and damage during the operation period. Depending on the size of the load, the data entered sequentially tended to be related to predictive accuracy, from relatively old low loads that required long-term memory to large loads that required new and short-term memory. Because the amount of data obtained by structural analysis is insufficient for learning Artificial Neural Networks, its usefulness was confirmed by investigating the possibility of utilizing the data that amplified the data calculated as a result of the structural analysis using the stacked autoencoder.

Dynamical prediction of two meteorological factors using the deep neural network and the long short-term memory (ΙΙ)

Shin Ki-Hong,Jung Jae-Won,Chang Ki-Ho,Kim Kyungsik,Jung Woon-Seon,Lee Dong-In,You Cheol-Hwan 한국물리학회 2022 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.80 No.12

This paper presents the predictive accuracy using two-variate meteorological factors, average temperature and average humidity, in neural network algorithms. We analyze result in fve learning architectures such as the traditional artifcial neural network, deep neural network, and extreme learning machine, long short-term memory, and long-short-term memory with peephole connections, after manipulating the computer simulation. Our neural network modes are trained on the daily time-series dataset during 7 years (from 2014 to 2020). From the trained results for 2500, 5000, and 7500 epochs, we obtain the predicted accuracies of the meteorological factors produced from outputs in ten metropolitan cities (Seoul, Daejeon, Daegu, Busan, Incheon, Gwangju, Pohang, Mokpo, Tongyeong, and Jeonju). The error statistics is found from the result of outputs, and we compare these values to each other after the manipulation of fve neural networks. As using the long-shortterm memory model in testing 1 (the average temperature predicted from the input layer with six input nodes), Tonyeong has the lowest root-mean-squared error (RMSE) value of 0.866 (%) in summer from the computer simulation to predict the temperature. To predict the humidity, the RMSE is shown the lowest value of 5.732 (%), when using the long short-term memory model in summer in Mokpo in testing 2 (the average humidity predicted from the input layer with six input nodes). Particularly, the long short-term memory model is found to be more accurate in forecasting daily levels than other neural network models in temperature and humidity forecastings. Our result may provide a computer simulation basis for the necessity of exploring and developing a novel neural network evaluation method in the future.

앵커 음성의 고저와 전달 속도가 뉴스 수용자에 미치는 영향

박덕춘(주저자) ( Dug Chun Park ) 커뮤니케이션디자인학회 2013 커뮤니케이션 디자인학연구 Vol.44 No.-

본 연구는 뉴스 앵커의 음성의 고저와 기사의 전달 속도가 앵커에 대한 수용자의 호감도, 신뢰도, 그리고 장단기 기억에 어떤 영향을 미치는지 실험을 통해 살펴보았다.본 연구를 위해 KBS 뉴스타임 영상물을 바탕으로 동일한 내용의 TV뉴스를 음성의 고저와 기사의 전달 속도에 따라 4가지 유형으로 제작하여, 대학생으로 구성된 피험자 집단에게 시청하게 한 후, 설문조사를 통해 뉴스 기사에 대한 장단기 기억과 앵커의 호감도와 신뢰도를 측정하여 이를 분석하였다. 분석결과, 저음의 앵커 음성에 노출된 집단이 고음의 앵커 음성에 노출된 집단보다 앵커에 대해 높은 호감도와 신뢰도를 보였다. 그리고 앵커의 기사 전달속도가 느린 뉴스에 노출된 집단이 전달 속도가 빠른 뉴스에 노출된 집단보다 장단기 기억을 더 잘하는 것으로 나타났다. 그러나 앵커 음성의 고저에 따른 수용자의 장단기 기억에는 유의미한 차이가 없었으며, 앵커의 전달 속도에 따라 수용자의 호감도, 신뢰도에도 의미있는 차이가 나타나지 않았다. This research explores the effect of high tone and low tone of news anchor on viewer``s favor, trust, long term and short term memory. For this experimental research, 4 different video clips made from KBS newstime video clip, were exposed to 4 university student groups and degree of favor, trust, long term and short term memory were measured through survey questions and analysed with SPSS program. This research found out that those students who watched TV news with anchor``s low tone showed higher favor and trust than those students who watched TV news with anchor``s high tone, and those students who watched TV news with low speed showed better long term and short term memory than those who watched TV news with high speed. However the effect of high tone and low tone of TV anchor on viewer``s long term and short term memory was not found, and the effect of anchor``s speed on viewer``s favor and trust was not found either.