• Korean Journal of Construction Engineering and Management
• /
• 제24권6호
• /
• pp.74-80
• /
• 2023

This study examines the characteristics of architectural review opinions as unstructured data, focusing on the most challenging risk for developers of small and medium-sized apartment projects in response to the increasing number of single-person households in Korea. Using multidimensional analysis methods, the study analyzes the review opinions of 25 projects in B City. Correspondence analysis and MDS (Multidimensional Scale) analysis show that, consistent with prior research, the keywords related to 'structure' and 'planning' dominate architectural review opinions in B City. While the MDS model's stress is very poor at 34.4%, correspondence analysis reveals that this is due to the characteristics of unstructured data in architectural reviews. In addition, the non-structured data analyzed in this study, such as architectural review opinions, exhibited a probability distribution with low kurtosis and high skewness, as they involved various combinations and occurrences of data depending on the discretion of the review committee members and the specific formats of different local governments. This often led to the emergence of keywords that differed significantly from commonly mentioned terms. Although the study has some limitations, it provides a foundation for future detailed analysis by identifying the characteristics of architectural review opinions as unstructured data.

• Journal of the Korean Society of Radiology
• /
• 제11권7호
• /
• pp.555-564
• /
• 2017

The current radiation risk assessment for occupational exposure is based on the measured exposure dose and health checkups of workers. This people-centered risk assessment may occur errors because absence of using personal dosimeter or unrelated health symptoms of individuals lead to difficulties in obtaining accurate data from workers. In addition, although the established legal upper dose limit was used as a reference for the assessment, it does not imply that this limit is the optimal dose of radiation workers should get; ALARA principle should always be appreciated. Therefore, a new risk assessment model that can take account of all the important factors and implement optimization of radiation protection is required at the national level. In this paper, based on the KOSHA Risk Assessment, we studied on the workplace-centered risk assessment model for radiation field rather than the people-centered. The result of the study derived a right model for radiation field through the analysis of the risk assessment methods in various fields and also found data acquisition methods and procedures for applying to the model. Multidimensional model centering on the workplace will enables more accurate radiation risk assessment by using a risk index and radar plot, and consequently contribute to the efficient worker management, preemptive worker protection and implementation of optimization of radiation protection.

• Nuclear Engineering and Technology
• /
• 제44권8호
• /
• pp.889-896
• /
• 2012

Recent progress in the computational fluid dynamics methods of two- and multiphase phase flows has already started opening up new exciting possibilities for using complete multidimensional models to simulate boiling systems. Combining this new theoretical and computational approach with novel experimental methods should dramatically improve both our understanding of the physics of boiling and the predictive capabilities of models at various scale levels. However, for the multidimensional modeling framework to become an effective predictive tool, it must be complemented with accurate mechanistic closure laws of local boiling mechanisms. Boiling heat transfer has been studied quite extensively before. However, it turns out that the prevailing approach to the analysis of experimental data for both pool boiling and forced-convection boiling has been associated with formulating correlations which normally included several adjustable coefficients rather than based on first principle models of the underlying physical phenomena. One reason for this has been the tendency (driven by practical applications and industrial needs) to formulate single expressions which encompass a broad range of conditions and fluids. This, in turn, makes it difficult to identify various specific factors which can be independently modeled for different situations. The objective of this paper is to present a mechanistic modeling concept for both pool boiling and forced-convection boiling. The proposed approach is based on theoretical first-principle concepts, and uses a minimal number of coefficients which require calibration against experimental data. The proposed models have been validated against experimental data for water and parametrically tested. Model predictions are shown for a broad range of conditions.

• Communications for Statistical Applications and Methods
• /
• 제18권2호
• /
• pp.171-177
• /
• 2011

In order to construct approximation functions for real data, it is necessary to remove the outliers from the measured raw data before constructing the model. Conventionally, visualization and maximum residual error have been used for outlier detection, but they often fail to detect outliers for nonlinear functions with multidimensional input. Although the standard support vector regression based outlier detection methods for nonlinear function with multidimensional input have achieved good performance, they have practical issues in computational cost and parameter adjustments. In this paper we propose a practical approach to outlier detection using support vector regression that reduces computational time and defines outlier threshold suitably. We apply this approach to real data examples for validity.

• Proceedings of the IEEK Conference
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.384-386
• /
• 2009

This paper introduces an effectively optimized Fuzzy model identification by means of complex and nonlinear system applying PSO algorithm. In other words, we use PSO(Particle Swarm Optimization) for identification of Fuzzy model structure and parameter. PSO is an algorithm that follows a collaborative population-based search model. Each particle of swarm flies around in a multidimensional search space looking for the optimal solution. Then, Particles adjust their position according to their own and their neighboring-particles experience. This paper identifies the premise part parameters and the consequence structures that have many effects on Fuzzy system based on PSO. In the premise parts of the rules, we use triangular. Finally we evaluate the Fuzzy model that is widely used in the standard model of gas data and sew data.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.253-256
• /
• 1996

Rainfall is one of the major and complicated elements of hydrologic system. Accurate prediction of rainfall is very important to mitigate storm damage. The neural network is a good model to be applied for the classification problem, large combinatorial optimization and nonlinear mapping. In this dissertation, rainfall predictions by the neural network theory were presented. A multi-layer neural network was constructed. The network learned continuous-valued input and output data. The network was used to predict rainfall. The online, multivariate, short term rainfall prediction is possible by means of the developed model. A multidimensional rainfall generation model is applied to Seoul metropolitan area in order to generate the 10-minute rainfall. Application of neural network to the generated rainfall shows good prediction. Also application of neural network to 1-hour real data in Seoul metropolitan area shows slightly good predictions.

• Asia pacific journal of information systems
• /
• 제32권3호
• /
• pp.601-629
• /
• 2022

Social networking sites (SNSs) have become an essential part of people's lives. It is thus crucial to understand how individuals use these platforms. Previous literature has divided usage into numerous activities and then grouped them into dimensions to avoid excessive granularity. However, these categories have not been derived from a uniform theoretical background; consequently, these dimensions are dispersed, overlapping, and disconnected from each other. This study argues that "SNS usage" is a complex phenomenon consisting of multiple activities that can be grouped into dimensions under the umbrella of communication theories and these dimensions are related to each other in a particular multi-dimensional architecture. "SNS usage" is conceptualized as a third-order construct formed by "producing," "consuming," and "communicating." "Producing," in turn, is proposed as a second-order construct manifested by "commenting," "general information sharing," and "self-disclosure." The proposed model was assessed with data collected from 414 USA adult users and PLS-SEM technique. The results show empirical support for the theorized model. SNS providers now have this architecture that clarifies the role of each dimension of use, which will allow them to design effective strategies to encourage the use of these networks.

• Wind and Structures
• /
• 제25권6호
• /
• pp.537-549
• /
• 2017

In this study, the turbulent flow around a bluff body for different wind velocities was investigated numerically by using its two- and three-dimensional models. These models were tested to verify the validity of the simulation by being compared with experimental results which were taken from the literature. Variations of non-dimensional velocities in different positions according to the bluff body height were analysed and illustrated graphically. When the velocity distributions were examined, it was seen that the results of both two- and three-dimensional models agree with the experimental data. It was also seen that the velocities obtained from two-dimensional model matched up with the experimental data from the ground to the top of the bluff body. Particularly, compared to the front part of the bluff body, results of the upper and back part of the bluff body are better. Moreover, after comparing the results from calculations by using different models with experimental data, the effect of multidimensional models on the obtained results have been analysed for different inlet velocities. The calculation results from the two-dimensional (2D) model are in satisfactory agreement with the calculation results of the three-dimensional model (3D) for various flow situations when comparing with the experimental data from the literature even though the 3D model gives better solutions.

• Journal of Korea Water Resources Association
• /
• 제56권7호
• /
• pp.439-449
• /
• 2023

Recent advancements in data measuring technology have facilitated the installation of various sensors, such as pressure meters and flow meters, to effectively assess the real-time conditions of water distribution systems (WDSs). However, as cities expand extensively, the factors that impact the reliability of measurements have become increasingly diverse. In particular, demand data, one of the most significant hydraulic variable in WDS, is challenging to be measured directly and is prone to missing values, making the development of accurate data generation models more important. Therefore, this paper proposes an adversarially trained autoencoder (ATAE) model based on generative deep learning techniques to accurately estimate demand data in WDSs. The proposed model utilizes two neural networks: a generative network and a discriminative network. The generative network generates demand data using the information provided from the measured pressure data, while the discriminative network evaluates the generated demand outputs and provides feedback to the generator to learn the distinctive features of the data. To validate its performance, the ATAE model is applied to a real distribution system in Austin, Texas, USA. The study analyzes the impact of data uncertainty by calculating the accuracy of ATAE's prediction results for varying levels of uncertainty in the demand and the pressure time series data. Additionally, the model's performance is evaluated by comparing the results for different data collection periods (low, average, and high demand hours) to assess its ability to generate demand data based on water consumption levels.

• The Journal of Information Technology and Database
• /
• 제6권2호
• /
• pp.115-130
• /
• 1999

The price structure of housing market varies depending upon market price policy rather than low or high price policy because of IMF. The object of this study is to develop an analysis system for analyzing housing market and its demand. The analysis system consists of four major categories: macro index analysis, market decision analysis, housing market analysis, and consumer analysis. We model each category by using a variety of techniques such as generalized linear model, categorical analysis, bubble analysis, drill-down analysis, price sensitivity meter analysis, optimum price index analysis, profit index measurement analysis, correspondence analysis, conjoint analysis, and multidimensional scaling analysis. Seoul apartment data is analyzed to demonstrate the practical usefulness of the system.