• 한국물환경학회지
• /
• 제24권6호
• /
• pp.784-792
• /
• 2008

Greater focus must be placed on ensuring that the water quality model (WQM) reflects the objective of its application and the characteristics of the water environment properly before it is selected. In the development or application of WQM, various factors influencing the model predictions should be reviewed so that it can perform more properly and reasonably based on scientific theory. This study reviewed the characteristic of existing WQM and the domestic river environment to find the requirements of the model application for TMDLs management in Korea. In this study, a water quality model, QUAL-NIER, was developed based on the USEPA's QUAL2E. The core structure and reaction scheme of the model was established followed by the formulation of equations according to the scheme with some supplements on the reaction mechanisms which are necessary for domestic rivers. Algorithms on the equations were set up and programmed to form a computer-based model. The developed model, QUAL-NIER was applied to the main stem of the Nakdong river. The model was calibrated and verified to data measured in 2004. The model results displayed good agrement with the field measurements for both calibration and verification. From this study, it was concluded that the developed QUAL-NIER model was very powerful with regard to the water quality simulation in domestic rivers.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 2000년도 학술발표회 발표논문집
• /
• pp.590-598
• /
• 2000

A multiplicative ARIMA model was tested and applied to analyze the periodicity and trends of 168 monthly oxygen demand data from the Noryanggin water quality gauging station in the downstream Han River. ARIMA model was identified to fit to the data using ACF and PACF tests, and the parameters estimated using an unconditional least square method. The residuals between the observed and forecasted data were acceptable with the Porte-Manteau test. A forecast of DO changes was made for its applications.

• 보건행정학회지
• /
• 제14권3호
• /
• pp.66-96
• /
• 2004

This study was intended to describe the hypertensive patient's lifestyle and quality of life by creating a hypothetic model on the lifestyle and quality of life and by examining a causeand effect relationship, and to contribute to countermeasures for practicing their lifestyle and improving the quality of life through creating a predictable model. Exogenous variable($\xi$) of hypothetic model in this study composed of a family support, hypertension knowledge, perceived benefit and toughness. Endogenous variable($\eta$) composed of self-esteem, perceived health state, depression, lifestyle and quality of life. There were 6 measured variables for exogenous variable(x). There were 9 measured variables(y) for endogenous variable. Also, there was error variable ($\delta,\;\epsilon$) of an individual. The survey was conducted for 207 hypertensive parents who received an out-patient service for 3 weeks from September 15, 2003 to October 3, 2003 after diagnosing as hypertension from 2 general hospitals in Daegu. As the conformance of hypothetic model in this study, there were $x^2$= 155.81, standard $x^2$ ($x^2$/df)=2.32, GFI=0.003, NFI=0.971, CFI=0.982, and RMSEA=0.080. Generally, the hypothetic model and actual data were well coincided. The higher the hypertension knowledge was(t=6.030), the higher the perceived benefit was(t=9.429), the higher the toughness was(t=2.783), and the higher the perceived health state was(t=2.282), the higher the lifestyle was. However, the degree of depression (t=-0.038), family support(t=1.161), and self-esteem(t=0.518) was not affected. The higher the family support was(t=10.476), the higher the self-esteem was(t=7.244), the higher the perceived health state was(t=6.996), the lower the degree of depression was(t=-2.044), and the higher the practice degree of lifestyle was(t=3.315), the higher the quality of life was. However, the toughness(t=1.672) didn't have a significant influence on the quality of life. It was modified to increase the model conformance and gain a conscious model As the result of model revision, for the model conformance, there were $x^2$= 118.43, standard $x^2$=1.69, GFI=0.923, NFI=0.976, CFI=0.982, and RMSEA=0.078. As the revised model showed the better conformance than hypothetic model, it seemed to be more suitable model. In the revised model, the perceived benefit(t=9.440) affected the lifestyle in the revised model. Then, the lifestyle was influenced by hypertension knowledge(t=6.139), toughness (t=2.757), family support(t=2.078), perceived health state(t=1.962) in the order. As a factor which affected the quality of life, there were the family support(t=l0.46l), self-esteem(t=7.368), perceived health state(t=6.989), lifestyle(t=3.316), toughness(t=2.584), and depression(t=-1.968) in the order. It showed the significant effect.

• 품질경영학회지
• /
• 제52권3호
• /
• pp.429-457
• /
• 2024

Purpose: This study aims to provide useful information to real estate investors by developing a profit determination model using artificial intelligence. The model analyzes the real estate markets of six selected cities from multiple perspectives, incorporating characteristics of the real estate market, economic indicators, and policies to determine potential profits. Methods: Data on real estate markets, economic indicators, and policies for five cities were collected and cleaned. The data was then normalized and split into training and testing sets. An AI model was developed using machine learning algorithms and trained with this data. The model was applied to the six cities, and its accuracy was evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared by comparing predicted profits to actual outcomes. Results: The profit determination model was successfully applied to the real estate markets of six cities, showing high accuracy and predictability in profit forecasts. The study provided valuable insights for real estate investors, demonstrating the model's utility for informed investment decisions. Conclusion: The study identified areas for future improvement, suggesting the integration of diverse data sources and advanced machine learning techniques to enhance predictive capabilities.

• 경영정보학연구
• /
• 제26권1호
• /
• pp.19-56
• /
• 2024

본 연구는 한국의 인공지능 학습용 데이터 구축 사업과 데이터의 공공 개방에 관한 정책 수행 기관, 데이터 구축 기업, 그리고 이를 활용하는 다양한 기관의 데이터 품질에 대해 이해를 제고하고, 신뢰할 수 있는 인공지능 알고리즘 개발에 있어 가장 중요한 학습용 데이터 품질에 대한 이론적 토대를 만들기 위한 실증적 연구이다. 이를 위해, 데이터의 속성 요인, 데이터 구축환경 요인, 데이터 타입 관련 요인 등 인공지능 학습용 데이터 품질과 관련된 중요 선행요인을 도입하여 이론적 모형을 제안한다. 본 연구는 393명의 인공지능 학습용 데이터 구축 기업과 인공지능 서비스 개발 기업의 실무 담당자를 대상으로 설문조사를 실시하여 데이터를 수집하였다. 데이터 분석은 퍼지셋 질적비교분석 방법과 인공신경망 분석을 통해 이루어졌으며, 분석 결과를 통해 인공지능 학습용 데이터 관련 학술적 및 실무적 시사점을 도출했다.

• 한국경영과학회지
• /
• 제42권1호
• /
• pp.53-63
• /
• 2017

Managing the quantitative measurement of the data control activities in enterprise wide is important to secure management of data governance. However, research on data governance is limited to concept definitions and components, and data governance research on evaluation models is lacking. In this study, we developed a model of quantitative assessment for data governance including the assessment area, evaluation index and evaluation matrix. We also, proposed a method of developing the model of quantitative assessment for data governance. For this purpose, we used previous studies and expert opinion analysis such as the Delphi technique, KJ method in this paper. This study contributes to literature by developing a quantitative evaluation model for data governance at the early stage of the study. This paper can be used for the base line data in objective evidence of performance in the companies and agencies of operating data governance.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.289-292
• /
• 2002

This paper provides a frame-layer method for controlling bit rate of compressed video data in real time. Our approach is easy to operate and can store encoded video data in real time without deteriorating the quality of an image. To provide ameliorated and consistent visual quality, a new concept named SOP (Set Of Pictures) and a new quantization parameter variation control algorithm based on a second-order rate-distortion model 〔2〕 are introduced. The total bit-budget is allocated efficiently to cope with unpredictable recording time by using the proposed algorithm and it is distributed to each frame. In the end, we show improved and consistent video quality with experimental results obtained from C-model of a MPEG-4 (simple-profile) encoder.

• 설비공학논문집
• /
• 제13권1호
• /
• pp.9-17
• /
• 2001

In this study, flow boiling experiments were performed using R-134a on a plain tube bundle. Tests were conducted for the following range of variables; quality from 0.1 to 0.9, mass flux from $8\;kg/m^2s$ to $26\;kg/m^2s$ and heat flux from $10\;kW/m^2s$ to $40\;kW/m^2s$. The heat transfer coefficients were strongly dependent on the heat flux. However, they were almost independent on the mass flux or quality. The data are compared with the modified Chen model, which satisfactorily () predicted the data. Original Chen model, however, did not adequately predict the effect of quality. The reason may be attributed to the flow pattern of the present test, where the bubbly flow prevailed for the entire test range. The heat transfer coefficients of the tube bundle were 6~40% higher than those of the single tube pool boiling.

• 품질경영학회지
• /
• 제30권4호
• /
• pp.106-119
• /
• 2002

The fatigue curve with estimated parameters represents the estimate of the median or mean life at a given applied stress But, in order to assist a designer in making decisions regarding the fatigue failure mode, it is common practice to construct a design curve on the lower or safe side of data. In this study, to overcome the limitations(i.e., no runout, equal variance, and quality of the approximation, etc) of Shen, Wirsching, and Cashman's method which suggested the approximate design curve for nonlinear models using tolerance interval constructed by Owen's method, an algorithm to find design curves under the fatigue model using a parametric bootstrap method, is proposed and illustrated with multiple fatigue data sets.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권2호
• /
• pp.7-13
• /
• 2021

Machine vision is a technology that helps the computer as if a person recognizes and determines things. In recent years, as advanced technologies such as optical systems, artificial intelligence and big data advanced in conventional machine vision system became more accurate quality inspection and it increases the manufacturing efficiency. In machine vision systems using deep learning, the image quality of the input image is very important. However, most images obtained in the industrial field for quality inspection typically contain noise. This noise is a major factor in the performance of the machine vision system. Therefore, in order to improve the performance of the machine vision system, it is necessary to eliminate the noise of the image. There are lots of research being done to remove noise from the image. In this paper, we propose an autoencoder based machine vision system to eliminate noise in the image. Through experiment proposed model showed better performance compared to the basic autoencoder model in denoising and image reconstruction capability for MNIST and fashion MNIST data sets.