• 서비스연구
• /
• 제13권2호
• /
• pp.99-113
• /
• 2023

고객 기반의 산업에서 고객 Retention은 기업의 경쟁력이라 할 수 있으며, 고객 Retention을 높이는 것은 기업의 경쟁력을 높이는 것이라 할 수 있다. 따라서, 미래 휴면 고객을 잘 예측하여 관리하는 것은 기업의 경쟁력을 높이는데 무엇보다 중요하다. 왜냐하면, 신규 고객을 유치하는데 필요한 비용이 기존 고객을 Lock-in 시키는데 드는 비용 보다 많은 것으로 알려져 있기 때문이다. 특히, 수 많은 카드사가 존재하는 국내 카드 산업의 휴면 카드를 관리하고자 정부에서 휴면 카드 자동 해지 제도를 도입하고 있으며, 카드 산업에서 휴면 고객을 관리하는 것이 무엇보다 중요한 과제로 떠오르고 있다. 본 연구에서는 카드 산업에서 휴면 고객을 예측하기 위해 Recurrent Neural Network (RNN)방법론을 사용하였으며, RNN방법론 중에서 긴 시간을 효율적으로 학습할 수 있는 Long-Short Term Memory (LSTM)을 활용하였다. 또한, 통합기술수용이론 (UTAUT)을 입각하여 카드 산업에서 휴면 고객을 예측하는데 필요한 변수를 재정의하였다. 그 결과 안정된 모형의 정확도와 F-1 score를 얻을 수 있었으며, Hit-Ratio를 통하여 모형의 안정된 결과를 입증하였다. 기존 연구에서 지적된 통합기술수용이론 (UTAUT)에서 발생 될 수 있는 인구통계학적 정보의 조절 효과도 발생 되지 않은 것을 보였으며, 이로 인해 통합기술수용이론(UTAUT)를 이용한 변수 선정 모형에서 LSTM을 이용한 휴면 고객 예측 모형은 편향되지 않고 안정된 결과를 가져다 줄 수 있다는 것을 입증하였다.

• 대한화학회지
• /
• 제46권5호
• /
• pp.456-465
• /
• 2002

이 연구의 목적은 눈에 보이지 않는 미시세계의 현상을 눈으로 볼수 있도록 애니메이션 자료를 개발하고 그 교육적 효과를 알아보는 것이다. 개발한 11개의 애니메이션 자료는 분자, 이온, 전자 등 입자 수준의 미시세계를 동영상으로 나타내었으며 교사가 ''전기화학''단원을 가르칠 때 사용할 수 있도록 하였다. 연구대상은 중학교 3학년 151명이었으며, 실험 집단과 통제 집단으로 나누어 애니메이션 자료를 이용한 수업과 전통적 수업을 16차시 동안 실시하였다. 처치 전에 학습자의 특성을 조사하기 위해 논리적 사고력 검사와 사전 개념 검사를 실시하였고 처치 후에는 사후 개념 검사, 과학에 관련된 태도 검사, 수업에 대한 인식 검사를 실시하였다. 그리고 4개월 후에 파지 효과를 알아보기 위해 사후 개념 검사를 다시 한번 실시하였다. 연구 결과에 의하면, 애니메이션 자료를 활용한 수업을 실시한 실험 집단이 개념 이해 검사에서 통제 집단보다 유의미하게 높은 점수를 얻었으며, 파지 검사와 태도 검사에서도 유의미하게 높은 점수를 얻었다. 이는 눈으로 볼 수 없는 미시세계에서 입자의 움직임을 보여준 수업 자료가 학생들의 개념이해와 학습 흥미도 증진에 효과적임을 의미하는 것이다.

• Nutrition Research and Practice
• /
• 제15권2호
• /
• pp.173-186
• /
• 2021

BACKGROUD/OBJECTIVES: Alzheimer's disease (AD) is the most common cause of dementia in the elderly. Due to the increased incidence of dementia, there is a corresponding increase concerning the importance of AD. In this study, we investigated the protective effects conferred by Zizyphus jujuba (Zj) and Zizyphus jujuba fermented by yeast (Zj-Y), on cognitive impairment in an AD mouse model. MATERIALS/METHODS: AD was induced by injecting amyloid beta25-35 (Aβ25-35) in ICR mice, and subsequently 200 mg/kg Zj or Zj-Y was administered daily for 14 days. The cognitive ability of AD mice was observed through behavioral experiments in T-maze, novel object recognition, and Morris water maze tests. We subsequently measured the levels of malondialdehyde (MDA), nitric oxide (NO), aspartate aminotransferase, and alanine aminotransferase in either tissues or serum. RESULTS: In behavioral tests, deterioration was revealed in the short- and long-term learning and memory functions in the Aβ25-35-injected control group compared to the normal group, indicating that Aβ25-35 injection impairs cognitive functions. However, administration of Zj and Zj-Y improved cognitive function in mice, as compared to the Aβ25-35-injected control mice. In addition, the Aβ25-35 induced elevations of MDA and NO in the brain, kidney, and liver were suppressed after exposure to Zj and Zj-Y. Especially, Zj-Y showed stronger scavenging effect against MDA and NO, as compared to Zj. CONCLUSIONS: Results of the present study indicate that Zj-Y exerts a protective effect on cognitive impairment and memory dysfunction, which is exerted by attenuating the oxidative stress induced by Aβ25-35.

• 한국항만경제학회지
• /
• 제27권3호
• /
• pp.41-64
• /
• 2011

이 연구는 케이프사이즈 시장과 파나막스 시장간의 비대칭 시간 가변 파급효과를 분석하기 위해서 조건부 평균에 전통적인 공적분항과 부분공적분항을 고려하고 있고, 조건부 분산에 레버리지 효과를 고려한 고정상관관계 GARCH와 동적상관관계 GARCH 모형을 이용하였다. 연구결과는 다음과 같다. 첫째, 두 시장간의 선-후행관계에 대해서 부분공적분항을 고려한 결과로부터 전기의 케이프사이즈 가격은 파나막스 시장가격에 유의적으로 정(+)의 영향을 미치고, 일반적인 공적분항을 고려하는 경우 두 시장간의 선-후행효과는 모두 유의적으로 정(+)의 효과를 갖는 것으로 나타났다. 둘째, 두 시장간의 장기적인 균형관계가 성립하지 않는 경우, 개별시장은 어떻게 반응하는가를 나타내는 오차항의 계수는 모두 통계적으로 유의적이고, 케이프사이즈 시장에서는 모두 음(-)의 값을 가지고 파나막스 시장의 경우에는 모두 정(+)의 값을 갖는 것으로 나타났다. 셋째, 두 시장간의 변동성의 파급효과에 대해서는 모든 모형에서 서로 영향을 주고 받는 것으로 나타났고, 통계적으로 유의하게 나타났다. 넷째, 레버리지 효과는 케이프사이즈 시장에서는 모두 유의적으로 정(+)의 값을 가지나, 파나막스 시장에서는 모두 통계적으로 유의적인 결과를 얻지 못하였다. 그러나 두 시장 모두에서 비대칭의 효과는 통계적으로 유의적인 것으로 나타나고 있다.

• Journal of Ginseng Research
• /
• 제38권1호
• /
• pp.1-7
• /
• 2014

Background: Although ginsenosides such as Rg1, Rb1 and Rg3 have shown promise as potential nutraceuticals for cognitive impairment, their use has been limited due to high production cost and low potency. In particular, the process of extracting pure Rg3 from ginseng is laborious and expensive. Methods: We described the methods in preparing ginseol k-g3, an Rg3-enriched fraction, and evaluated its effects on scopolamine-induced memory impairment in mice. Results: Ginseol k-g3 (25-200 mg/kg) significantly reversed scopolamine-induced cognitive impairment in the passive avoidance, but not in Y-maze testing. Ginseol k-g3 (50 and 200 mg/kg) improved escape latency in training trials and increased swimming times within the target zone of the Morris water maze. The effect of ginseol k-g3 on the water maze task was more potent than that of Rg3 or Red ginseng. Acute or subchronic (6 d) treatment of ginseol k-g3 did not alter normal locomotor activity of mice in an open field. Ginseol k-g3 did not inhibit acetylcholinesterase activity, unlike donezepil, an acetylcholinesterase inhibitor. Rg3 enrichment through the ginseol k-g3 fraction enhanced the efficacy of Rg3 in scopolamine-induced memory impairment in mice as demonstrated in the Morris water maze task. Conclusion: The effects of ginseol k-g3 in ameliorating scopolamine-induced memory impairment in the passive avoidance and Morris water maze tests indicate its specific influence on reference or long-term memory. The mechanism underlying the reversal of scopolamine-induced amnesia by ginseol k-g3 is not yet known, but is not related to anticholinesterase-like activity.

• Journal of Acupuncture Research
• /
• 제33권4호
• /
• pp.49-63
• /
• 2016

Objectives : This research was performed to investigate the effects of Jodeungsan pharmacopuncture(PA-J) of focal brain ischemia induced by middle cerebral artery occlusion(MCAO) in rats. Methods : The subjects were divided into 4 groups : control, acupuncture, pharmacopuncture PA-J1(11.43 mg / 250 g / $40{\mu}{\ell}$) and pharmacopuncture PA-J2(2.29 mg / 250 g / $40{\mu}{\ell}$). The focal brain ischemia was induced by intraluminal filament insertion into the middle cerebral artery. After 3 days of MCAO, Jodeungsan pharmacopuncture treatment was performed on the GB20, and the day after being treated with pharmacopuncture, the Morris water maze test was carried out on the assigned group. The series of processes were administered 6 times. Thereafter mGluR5, density of neuronal cell and ChAT were measured. Results : The results were as follows. 1. The distance to target significantly decreased in the 2nd trial of the Acu group on the water maze test for short-term memory. 2. The distance to target significantly decreased in the 4th trial of the PA-J2 group on the water maze test for long-term memory. 3. The intensity of mGluR5 significantly increased in the PA-J1 group compared with the control group. 4. The neuroprotective effect on the hippocampal CA1 significantly increased in the PA-J1 and PA-J2 groups compared with the control group. 5. The density of ChAT in the hippocampal CA1 significantly increased in the PA-J1 and PA-J2 groups compared with the control group. Conclusion : These results suggest that Jodeungsan pharmacopuncture may improve memory and cognitive impairment and also have neuroprotective effects on focal brain ischemia.

• Structural Monitoring and Maintenance
• /
• 제7권2호
• /
• pp.109-124
• /
• 2020

Benefiting from the massive monitoring data collected by the Structural health monitoring (SHM) system, scholars can grasp the complex environmental effects and structural state during structure operation. However, the monitoring data is often missing due to sensor faults and other reasons. It is necessary to study the recovery method of missing monitoring data. Taking the structural temperature monitoring data of Nanjing Dashengguan Yangtze River Bridge as an example, the long short-term memory (LSTM) network-based recovery method for missing structural temperature data is proposed in this paper. Firstly, the prediction results of temperature data using LSTM network, support vector machine (SVM), and wavelet neural network (WNN) are compared to verify the accuracy advantage of LSTM network in predicting time series data (such as structural temperature). Secondly, the application of LSTM network in the recovery of missing structural temperature data is discussed in detail. The results show that: the LSTM network can effectively recover the missing structural temperature data; incorporating more intact sensor data as input will further improve the recovery effect of missing data; selecting the sensor data which has a higher correlation coefficient with the data we want to recover as the input can achieve higher accuracy.

• Toxicological Research
• /
• 제17권
• /
• pp.257-261
• /
• 2001

Recently, a traditional medicine called Gouteng-san, which consists of eleven herbs, was reported to be effective in treating vascular dementia with a double-blind, placebo-controlled study. Gout-eng-san is also used for patients with vascular dementia in combination with Si-Wu-Tang. The effect of Gouteng-san and Si-Wu-Tang on deficit of learning behavior was investigated using step-down passive avoidance task in mice. Hot-water extract of Gouteng-san (1.5 and 6 g/kg, p.o.) significantly prolonged the step-down latency shortened by scopolamine. The extract of Uncaria hook (150 mg/kg, p.o.), one of the component herb of Gouteng-san, significantly prevented the decrease in the latency after scopolamine. Hot-water extract of Si-Wu-Tang (1.5 and 6 g/kg of dried herbs, p.o.) prevented dose-dependently scopola-mine-induced disruption qf learning behavior. Si-Wu-Tang also prevented the ischemia-induced deficit of learning behavior. Both hot water extract of peony and angelica (1.5 g/kg, p.o.), which are component herbs qf Si-Wu-Tang, prevented the scopolamine-induced learning behavior deficit. Scopolamine (10 uM) suppressed long-term potentiation (LTP) of population spike in the CA1 region of the rat hippocampal slices. Peoniflorin (0.1~ 1uM) extracted from paeony root significantly ameliorated scopolamine-induced inhibition of LTR These results suggest that improvement of deficit of learning behavior by Gouteng-san and Si-Wu-Tang is mediated by direct and/or indirect activation of the cholinergic system in the brain.

• Smart Structures and Systems
• /
• 제24권4호
• /
• pp.459-474
• /
• 2019

This study develops a novel recursive algorithm to significantly enhance the computation efficiency of a recently proposed stochastic subspace identification (SSI) methodology based on an alternative stabilization diagram. Exemplified by the measurements taken from the two investigated office buildings, it is first demonstrated that merely one sixth of computation time and one fifth of computer memory are required with the new recursive algorithm. Such a progress would enable the realization of on-line and almost real-time monitoring for these two steel framed structures. This recursive SSI algorithm is further applied to analyze 20 months of monitoring data and comprehensively assess the environmental effects. It is certified that the root-mean-square (RMS) response can be utilized as an excellent index to represent most of the environmental effects and its variation strongly correlates with that of the modal frequency. More detailed examination by comparing the monthly correlation coefficient discloses that larger variations in modal frequency induced by greater RMS responses would typically lead to a higher correlation.

• ETRI Journal
• /
• 제41권3호
• /
• 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.