• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.1
• /
• pp.81-100
• /
• 1996

본 연구에서는 보다 효과적인 기업신용평가를 위하여, 통계적 방법과 인공지능 방법을 결합한 결합모형을 제시햐고자 한다. 이를 위하여 본 연ㄴ구에서는 통계적인 모형중에서 가장 널리 활용되고 있는 MDA (Multivariate Discriminant Analysis) 와 인공지능적인 방법으로서 최근에 널리 사용되고 있는 인공싱경망( neural network)모형을 휴리스틱한 방법으로 결합한다. 이러한 결합모형의 성과를 증명하기 위하여 우리나라의 대표적인 3대 기업신용평가 기관에서 수집한 1043개의 기업신용평가자료를 기초로 실혐을 하고, 그 결과를 기존의 MDA 및 인공신경망 방법에 의한 결과와 비교하였다. 실험결과, 통계적으로도 유의하고, 실무적인 관점에서도 의미가 있는 기업신용펑가 결과를 유도할 수 있었다.

• Journal of Korea Water Resources Association
• /
• v.45 no.12
• /
• pp.1321-1330
• /
• 2012

In this study, the daily inflow at the basin of Chungju dam is predicted using wavelet-artificial neural network for nonlinear model. Time series generally consists of a linear combination of trend, periodicity and stochastic component. However, when framing time series model through these data, trend and periodicity component have to be removed. Wavelet transform which is denoising technique is applied to remove nonlinear dynamic noise such as trend and periodicity included in hydrometeorological data and simple noise that arises in the measurement process. The wavelet-artificial neural network (WANN) using data applied wavelet transform as input variable and the artificial neural network (ANN) using only raw data are compared. As a results, coefficient of determination and the slope through linear regression show that WANN is higher than ANN by 0.031 and 0.0115 respectively. And RMSE and RRMSE of WANN are smaller than those of ANN by 37.388 and 0.099 respectively. Therefore, WANN model applied in this study shows more accurate results than ANN and application of denoising technique through wavelet transforms is expected that more accurate predictions than the use of raw data with noise.

• Journal of Intelligence and Information Systems
• /
• v.15 no.3
• /
• pp.151-165
• /
• 2009

Fuzzy neural network is an integrated model of artificial neural network and fuzzy system and it has been successfully applied in control and forecasting area. Recently ANFIS(Adaptive Network-based Fuzzy Inference System) has been noticed widely among various fuzzy neural network models because of its outstanding accuracy of control and forecasting area. We design a new classification model based on ANFIS and evaluate it in terms of classification accuracy. We identified ANFIS-based classification model has higher classification accuracy compared to existing classification model, C5.0 decision tree model by comparing their experimental results.

• Land and Housing Review
• /
• v.14 no.1
• /
• pp.67-82
• /
• 2023

Using the Artificial Neural Network model and Binary Logistic Regression model, this study investigates influence factors on the quality of life in terms of housing environment during the COVID-19 in Seoul. The results show that the lower the satisfaction level of housing policy, the lower the quality of life in the employment field and the lower the quality of residential field. On the other hand, permanent workers and self-employed respondents have experienced improvement in residential quality during the pandemic. A limitation of this study is associated with disentangling the causal relationship using the 'black box' characteristics of ANN method.

• Journal of Digital Convergence
• /
• v.16 no.1
• /
• pp.231-242
• /
• 2018

The efficient management of the Length of Stay(LOS) is important in hospital. It is import to reduce medical cost for patients and increase profitability for hospitals. In order to efficiently manage LOS, it is necessary to develop an artificial intelligence-based prediction model that supports hospitals in benchmarking and reduction ways of LOS. In order to develop a predictive model of LOS for acute stroke patients, acute stroke patients were extracted from 2013 and 2014 discharge injury patient data. The data for analysis was classified as 60% for training and 40% for evaluation. In the model development, we used traditional regression technique such as multiple regression analysis method, artificial intelligence technique such as interactive decision tree, neural network technique, and ensemble technique which integrate all. Model evaluation used Root ASE (Absolute error) index. They were 23.7 by multiple regression, 23.7 by interactive decision tree, 22.7 by neural network and 22.7 by esemble technique. As a result of model evaluation, neural network technique which is artificial intelligence technique was found to be superior. Through this, the utility of artificial intelligence has been proved in the development of the prediction LOS model. In the future, it is necessary to continue research on how to utilize artificial intelligence techniques more effectively in the development of LOS prediction model.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.6
• /
• pp.1189-1198
• /
• 2015

For the effects of the global financial crisis cause hedge funds to have a strong influence on financial markets, it is needed to study new approach method to predict hedge fund survival. This paper proposes to organize fuzzy neural network using hedge fund data as input to predict hedge fund survival. The variables of hedge fund data are ambiguous to analyze and have internal uncertainty and these characteristics make it challenging to predict their survival from the past records. The object of this study is to evaluate the predictability of fuzzy neural network which uses grades of membership to predict survival. The results of this study show that proposed system is effective to predict the hedge funds survival and can be a desirable solution which helps investors to support decision-making.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1B
• /
• pp.39-50
• /
• 2006

The flood control offices of main rivers have used a storage function model to forecast flood stage in Korea and studies of flood forecasting actively have been done even now. On this account, the storage function model, which is used in flood control office, regression models and artificial neural network model are applied into flood forecasting of study watershed in this paper. The result obtained by each method are analyzed for the comparative study. In case of storage function model, this paper uses the representative parameters of the flood control offices and the optimized parameters. Regression coefficients are obtained by regression analysis and neural network is trained by backpropagation algorithm after selecting four events between 1995 to 2001. As a result of this study, it is shown that the optimized parameters are superior to the representative parameters for flood forecasting. The results obtained by multiple, robust, stepwise regression analysis, one of the regression methods, show very good forecasts. Although the artificial neural network model shows less exact results than the regression model, it can be efficient way to produce a good forecasts.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.6
• /
• pp.981-987
• /
• 2017

Jeju Island is a volcanic island which has a large permeability. Groundwater is a major water resources and its proper management is essential. Especially, there is a multilevel restriction due to the groundwater level decline during a drought period to protect sea water intrusion. Preliminary countermeasure using long-term groundwater level prediction is necessary to use agricultural groundwater properly. For this purpose, the monthly groundwater level prediction technique by Artificial Neural Network model was developed and applied to the representative monitoring wells. The monthly prediction model showed excellent results for training and test periods. The continuous groundwater level prediction model also developed, which used the monthly forecasted values adaptively as input data. The characteristics of groundwater declines were analyzed under extreme cases without precipitation for several months.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.1-4
• /
• 2001

The classification performance of Artificial Neural Network (ANN) and RBF-NN was compared for Landsat TM image. The RBF-NN was validated for three unique landuse types (e.g. Mixed landuse area, Cultivated area, Urban area), different input band combinations and classification class. The bootstrap resampling technique was employed to estimate the confidence intervals and distribution for unit load, The pollutant generation was varied significantly according to the classification accuracy and percentile unit load applied. Especially in urban area, where mixed landuse is dominant, the difference of estimated pollutant load is largely varied.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.216-216
• /
• 2021

해저지진, 해저붕괴 및 해저화산분출 등에 발생되는 지진해일은 파장이 수십에서 수백 km에 이르는 장파로서 에너지 손실없이 먼 거리를 전파할 수 있으며, 수심이 상대적으로 얕은 해안가에 도달하면 범람에 의해 인명 및 재산피해를 야기시킬 수 있다. 예를 들어, 2004년 12월 26일에 발생한 수마트라 지진해일은 약 30만명의 인명피해와 약 10조원의 재산피해를 가져왔으며, 2011년 3월 11일에 발생한 동일본 지진해일은 약 2만명의 인명피해와 약 330조의 재산피해를 유발시켰다. 더욱이, 지진해일에 의해 폭발한 후쿠시마 원자력발전소에서의 방사능 유출은 10년이 지난 현재도 생태계 교란, 방사능 피폭 등의 피해를 일으키고 있다. 우리나라도 1983년 5월 26일 발생한 동해 중부지진해일에 의해 삼척시 임원항 및 인근에서 인명피해(1명 사망, 2명 실종)와 약 2억원의 재산피해가 발생하였다. 최근, 4차 산업혁명으로서 빅데이터를 기반으로 한 다양한 인공지능기술이 개발되고 있으며, 많은 분야에서 이 기술을 적용하고자 노력하고 있다. 특히, 과학 및 공학분야에서도 이를 융합하는 연구 및 활용하는 사례가 증가하고 있다. 본 연구에서는 1983년 발생한 중부지진해일에 의해 인명 및 재산피해가 발생한 임원항을 대상으로 지진해일 수치모형실험을 수행하며, 수치모형실험 결과를 토대로 인공지능 모델 중 합성신경망 (Convolution Neural Network)을 활용하여 인공지능을 통한 지진해일 범람구역을 산정 및 평가하고자 한다.