• International Journal of Computer Science & Network Security
• /
• 제21권3호
• /
• pp.103-111
• /
• 2021

Obesity affects individuals of all gender and ages worldwide; consequently, several studies have performed great works to define factors causing it. This study develops an effective method to trace obesity levels based on supervised data mining techniques such as Random Forest and Multi-Layer Perception (MLP), so as to tackle this universal epidemic. Notably, the dataset was from countries like Mexico, Peru, and Colombia in the 14- 61year age group, with varying eating habits and physical conditions. The data includes 2111 instances and 17 attributes labelled using NObesity, which facilitates categorization of data using Overweight Levels l I and II, Insufficient Weight, Normal Weight, as well as Obesity Type I to III. This study found that the highest accuracy was achieved by Random Forest algorithm in comparison to the MLP algorithm, with an overall classification rate of 96.7%.

• 한국정밀공학회지
• /
• 제12권6호
• /
• pp.120-127
• /
• 1995

A nonlinear empirical state-space model of the Artificial Neural Network(ANN) has been developed. The nonlinear model structure incorporates characteristic, so as to enable identification of the transient response, as well as the steady-state response of a dynamic system. A hybrid feedfoward/feedback neural network, namely a Local Time Delayed Recurrent Multi-layer Perception(RMLP), is the model structure developed in this paper. RMLP is used to identify nonlinear dynamic system in an input/output sense. The feedfoward protion of the network architecture provides with the well-known curve fitting factor, while local recurrent and cross-talk connections provides the dynamics of the system. A dynamic learning algorithm is used to train the proposed network in a supervised manner. The derived dynamic learning algorithm exhibit a computationally desirable characteristic; both network sweep involved in the algorithm are performed forward, enhancing its parallel implementation. RMLP state-space and its associate learning algorithm is demonstrated through a simple examples. The simulation results are very encouraging.

• 전자공학회논문지CI
• /
• 제41권3호
• /
• pp.67-77
• /
• 2004

최근 잡음환경에서 신뢰도 높은 음성인식을 위해 음성정보와 영상정보를 융합하는 방법이 활발히 연구되고 있다. 본 논문에서는 이절적인 정보의 융합에 적합한 신경망 모델을 기반으로 음성, 영상 및 문맥 정보 등 다양한 정보를 융합하여 잡음 환경에서 고려단어를 인식하는 음성인식 기법에 대하여 기술한다. 음성과 영상 특징을 이용한 이중 모드 신경망 BMNN(BiModal Neural Network)을 제안한다. BMM은 4개 층으로 이루어진 다층퍼셉트론의 구조를 가지며 각 층은 입력 특징의 추상화 기능을 수행한다. BMNN에서는 제 3층이 잡음에 의한 음성 정보의 손실을 보상하기 위하여 음성과 영상 특징을 통합하는 기능을 수행한다. 또한, 잡음환경에서 음성 인식률을 향상시키기 위해 사용자가 말한 단어들의 순차 패턴을 나타내는 문맥정보를 이용한 후처리 방법을 제안한다. 잡음환경에서 BMNN은 단순히 음성만을 사용한 것 보다 높은 성능을 보임으로써 그 타당성을 확인할 수 있을 뿐 아니라, 특히 문맥을 이용한 후처리를 하였을 경우 잡음 환경에서 90%이상의 인식률을 달성하였다 본 연구는 잡음환경에서 강인한 음성인식을 위해 다양한 추가 정보를 사용함으로써 성능을 향상시킬 수 있음을 제시한다.

• 한국실내디자인학회논문집
• /
• 제23권2호
• /
• pp.166-173
• /
• 2014

This study analyzed visual-perceptual centrality as a measure to solve the confusion that might occur in visitor's movement in analyzing contemporary museums that have recently expand. Also, this study proposed a new methodology to understand the relationships of the occurrence and convergence of visual-perceptual power. The existing visibility graph analysis mostly ends up in identifying nodes corresponding to the physical environment of museums while the newly proposed visibility graph analysis is a methodology that can analyze the landmark on which the visual-perceptual power occurs and converges, by setting visible space elements as analysis variables, which showed a possibility to draw more reliable results in parallel to the existing analysis, and the results of this analysis are as follows. A public museum located in downtown has a limitation that it cannot but be multi-layered as a way to enlarge for the accommodation of a variety of programs on a limited site. Also, this study found that the problem of visitor orientation occurring in the multi-layer process was solved by a major space which was visual-perceptually void. Also, its types can be summarized into two types: One is setting parameter spaces (ramps, staircases, hallways) linked visual-perceptually to the major space and the other is solving the confusion of the visitor's movement by composing a simple exhibition space. In particular, it is judged that the type solving the confusion of the visitor's movement by composing a simple exhibition space also had a relationship with changes in the characteristics of the exhibited works by period. In addition, it was found that the entrance also appeared in two types: dependent type and independent type in the major space according to the characteristics of the major space.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1999년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.262-266
• /
• 1999

Several commercial companies that plan to provide improved panchromatic and/or multi-spectral remote sensor data in the near future are suggesting that merge datasets will be of significant value. This study evaluated the utility of one major merging process-process components analysis and its inverse. The 6 bands of 30$\times$30m Landsat TM data and the 10$\times$l0m SPOT panchromatic data were used to create a new 10$\times$10m merged data file. For the image classification, 6 bands that is 1st, 2nd, 3rd, 4th, 5th and 7th band may be used in conjunction with supervised classification algorithms except band 6. One of the 7 bands is Band 6 that records thermal IR energy and is rarely used because of its coarse spatial resolution (120m) except being employed in thermal mapping. Because SPOT panchromatic has high resolution it makes 10$\times$10m SPOT panchromatic data be used to classify for the detailed classification. SPOT as the Landsat has acquired hundreds of thousands of images in digital format that are commercially available and are used by scientists in different fields. After the merged, the classifications used supervised classification and neural network. The method of the supervised classification is what used parallelepiped and/or minimum distance and MLC(Maximum Likelihood Classification) The back-propagation in the multi-layer perception is one of the neural network. The used method in this paper is MLC(Maximum Likelihood Classification) of the supervised classification and the back-propagation of the neural network. Later in this research SPOT systems and images are compared with these classification. A comparative analysis of the classifications from the TM and merged SPOT/TM datasets will be resulted in some conclusions.

• Wind and Structures
• /
• 제2권1호
• /
• pp.25-40
• /
• 1999

In this paper, artificial neural networks, a new kind of intelligent method, are employed to model and predict amplitude dependent damping in buildings based on our full-scale measurements of buildings. The modelling method and procedure using neural networks to model the damping are studied. Comparative analysis of different neural network models of damping, which includes multi-layer perception network (MLP), recurrent neural network, and general regression neural network (GRNN), is performed and discussed in detail. The performances of the models are evaluated and discussed by tests and predictions including self-test, "one-lag" prediction and "multi-lag" prediction of the damping values at high amplitude levels. The established models of damping are used to predict the damping in the following three ways : (1) the model is established by part of the data measured from one building and is used to predict the another part of damping values which are always difficult to obtain from field measurements : the values at the high amplitude level. (2) The model is established by the damping data measured from one building and is used to predict the variation curve of damping for another building. And (3) the model is established by the data measured from more than one buildings and is used to predict the variation curve of damping for another building. The prediction results are discussed.

• 한국컴퓨터정보학회논문지
• /
• 제16권8호
• /
• pp.137-146
• /
• 2011

에너지 소모에 대한 증가되는 소비자의 관심을 지원하기 위하여 가전기기의 에너지 모니터링과 예측 방식을 제안한다. 제안한 시스템은 0.5초마다 전류 센서를 지나가는 전류량을 측정하는 스마트 플러그라는 일반 전기 콘센트로 설계하고, 신경회로망의 훈련과 시험 데이터를 얻기 위해 평균기온, 최저기온, 초고기온, 습도, 일조시간의 날씨 정보를 입력 데이터로 사용하고, 스마트 플러그를 통한 전기 사용량을 목표값으로 사용하였다. 훈련을 위한 실험데이터를 사용하여 역전파 알고리즘을 기반으로 한 신경회로망을 구성하였다. 입력과 출력 데이터의 비선형 매핑을 위해 다층신경회로망을 사용하였다. 제안한 신경회로망 모델은 상관관계 계수가 0.9965로 우수하게 전기 사용량을 예측할 수 있는 것을 확인하였으며, 예측의 평균 제곱 오차는 0.02033이다.

• 한국정보통신학회논문지
• /
• 제21권9호
• /
• pp.1718-1724
• /
• 2017

최근 IoT기반으로 가전제품을 실시간 모니터링을 하는 스마트 플러그가 활성화 되고 있다. 이를 통해 상시 실시간 에너지 소비 모니터링을 통한 소비자의 에너지 절약 유도를 하고, 소비자 설정 기반의 알람 기능을 통해 소비전력을 절감하는 효과를 보고 있다. 본 논문에서는 이러한 실시간 모니터링을 위해 벽 전원 콘센트에서 나오는 교류 전류를 측정한다. 이때, 가전제품마다의 전류 패턴을 분류하고 어떤 제품이 동작하는지 판단을 위해 딥러닝(Deep learning)으로 실험하였다. 전류 패턴의 학습으로 제품의 종류에 따른 인식 성능을 검증하기 위하여, 교차 검증 방법과 붓스트랩(Bootstrap) 검증 방법을 이용하였다. 또한 Cost function과 학습 성공률(Accuracy)이 Train 데이터와 Test 데이터가 동일함을 확인하였다.

• 생물정신의학
• /
• 제27권1호
• /
• pp.9-17
• /
• 2020

Objectives The purpose of this study is to investigate the interest level, perception and attitude of suicide high-risk group toward the media's suicide report, and also to provide the fundamental data in order to strengthen both the standard and the recommendation of media coverage in Korea. Methods We analyzed the data from the 2013 The Korea National Suicide Survey. The study included 1500 participants aged between 19 and 75 years. The participants were selected through the regional multi-layer stratification method using the sampling frame of the 2010 Census. One-on-one face-to-face interviews investigated the interest level, opinions, and attitudes toward suicide reports and collected data on past suicide history and demographic data. The collected data were evaluated using multiple logistic regression analysis. A probability level of p < 0.05 was considered statistically significant. Results The suicide high risk groups with histories of suicidal ideations or attempts showed more interest in suicidal reports [suicidal thought odds ratio (OR) = 1.81, 95% confidence interval (CI) = 1.40-2.34 ; suicidal attempt OR = 3.21, 95% Cl = 1.52-6.78]. However, there was no difference among the groups in thought that suicide reports incite suicide (suicidal thought OR = 1.26, 95% Cl = 0.92-1.73 ; suicidal attempt OR = 0.96, 95% Cl = 0.44-2.09). The suicide high risk groups showed a positive attitude toward suicide reports, but it was not statistically significant (suicidal thought OR = 1.10, 95% Cl = 0.41-2.92 ; suicidal attempt OR = 1.10, 95% Cl = 0.76-1.59). Conclusions We suggest the need of careful attitude of media suicidal reports not to bring about copycat suicides. and also we offer suicide prevention campaign based on media leverage.

• 지능정보연구
• /
• 제24권4호
• /
• pp.67-83
• /
• 2018

2013년 누적인원 2억명을 돌파한 한국의 영화 산업은 매년 괄목할만한 성장을 거듭하여 왔다. 하지만 2015년을 기점으로 한국의 영화 산업은 저성장 시대로 접어들어, 2016년에는 마이너스 성장을 기록하였다. 영화산업을 이루고 있는 각 이해당사자(제작사, 배급사, 극장주 등)들은 개봉 영화에 대한 시장의 반응을 예측하고 탄력적으로 대응하는 전략을 수립해 시장의 이익을 극대화하려고 한다. 이에 본 연구는 개봉 후 역동적으로 변화하는 관람객 수요 변화에 대한 탄력적인 대응을 할 수 있도록 주차 별 관람객 수를 예측하는데 목적을 두고 있다. 분석을 위해 선행연구에서 사용되었던 요인 뿐 아니라 개봉 후 역동적으로 변화하는 영화의 흥행순위, 매출 점유율, 흥행순위 변동 폭 등 선행연구에서 사용되지 않았던 데이터들을 새로운 요인으로 사용하고 Naive Bays, Random Forest, Support Vector Machine, Multi Layer Perception등의 기계학습 기법을 이용하여 개봉 일 후, 개봉 1주 후, 개봉 2주 후 시점에는 차주 누적 관람객 수를 예측하고 개봉 3주 후 시점에는 총 관람객 수를 예측하였다. 새롭게 제시한 변수들을 포함한 모델과 포함하지 않은 모델을 구성하여 실험하였고 비교를 위해 매 예측시점마다 동일한 예측 요인을 사용하여 총 관람객 수도 예측해보았다. 분석결과 동일한 시점에 총 관람객 수를 예측했을 경우 보다 차주 누적 관람객 수를 예측하는 것이 더 높은 정확도를 보였으며, 새롭게 제시한 변수들을 포함한 모델의 정확도가 대부분 높았으며 통계적으로 그 차이가 유의함으로써 정확도에 기여했음을 확인할 수 있었다. 기계학습 기법 중에는 Random Forest가 가장 높은 정확도를 보였다.