• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.162-164
• /
• 2003

The multi-disciplinary research project Strengthening Local Authorities in Risk Management (SLARIM), initiated by ITC, includes three case study cities in Asia. An important question is: what are the essential data for risk management and how to access such data. The role of common sources (e.g. census data), data derived from remote sensing (high-resolution satellite imagery, aerial photos), and data from close sensing (field observation, including mobile GIS) to acquire essential risk management data will be discussed. Special attention is given to the question of the minimum area and to disaggregating population data. A few examples are given of Kathmandu / Lalitpur, Nepal.

• Asian Journal for Public Opinion Research
• /
• v.3 no.4
• /
• pp.156-175
• /
• 2016

This paper considers whether random sampling always produces more accurate survey results in the case of South Korea. We compare information from the 2010 census to the demographic variables of three public opinion surveys from South Korea: Gallup Korea's Omnibus Survey (Survey A) is conducted every two months by Gallup Korea; the annual Social Survey (Survey B) is conducted by Statistics Korea (KOSTAT); the Korean General Social Survey (KGSS or Survey C) is conducted annually by the Survey Research Center (SRC) at Sungkyunkwan University (SKKU). Survey A uses quota sampling after randomly selecting the neighborhood and initial addresses; Survey B uses random sampling, but allows replacements in some situations; Survey C uses simple random sampling. Data from more than one year was used for each survey. Our analysis suggests that Survey B is the most representative in most respects, and, in some respects, Survey A may be more representative than Survey C. Data from Survey C was the least stable in terms of representativeness by geographical area and age. Single-person households were underrepresented in both Surveys A and C, but the problem was more severe in Survey A. Four-person households and married persons were both over-represented in Survey A. Less educated people were under-represented in both Survey A and Survey C. There were differences in income level between Survey A and Survey C, but income data was not available for Survey B or the census, so it is difficult to ascertain which survey was more representative in this case.

• Proceedings of the KSRS Conference
• /
• 1999.11a
• /
• pp.63-69
• /
• 1999

The purpose of this study is to develop Water Quality Management System(WQMS), which performs calculating pollutant discharge and forecasting water quality with water pollution model. Operational water quality management requires not only controlling pollutants but acquiring and managing exact information. A GIS software, ArcView was used to enter or edit geographic data and attribute data, and MapObject was used to customize the user interface. PCI, a remote sensing software, was used for deriving land cover classification from 20 m resolution SPOT data by image processing. WQMS has two subsystems, Database Subsystem and Modelling subsystem. Database subsystem consisted of watershed data from digital map, remote sensing data, government reports, census data and so on. Modelling subsystem consisted of NSPLM(NonStorm Pollutant Load Model)-SPLM(Storm Pollutant Load Model). It calculates the amount of pollutant and predicts water quality. This two subsystem was connected through graphic display module. This system has been calibrated and verified by applying to Mokhyun stream watershed.

• Journal of Korean Institute of Industrial Engineers
• /
• v.30 no.4
• /
• pp.267-276
• /
• 2004

Much of previous attention on researches of the decision tree focuses on the splitting criteria and optimization of tree size. Nowadays the quantity of the data increase and relation of variables becomes very complex. And hence, this comes to have plenty number of unnecessary node and leaf. Consequently the confidence of the explanation and forecasting of the decision tree falls off. In this research report, we propose some decision tree algorithms considering the interaction of predictor variables. A generic algorithm, the k-1 Algorithm, dealing with the interaction with a combination of all predictor variable is presented. And then, the extended version k-k Algorithm which considers with the interaction every k-depth with a combination of some predictor variables. Also, we present an improved algorithm by introducing control parameter to the algorithms. The algorithms are tested by real field credit card data, census data, bank data, etc.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.4
• /
• pp.116-128
• /
• 2011

In this paper, we propose the structure of a high-performance face-detection engine that responds well to facial rotating changes using rotation transformation which minimize the required memory usage compared to the previous face-detection engine. The validity of the proposed structure has been verified through the implementation of FPGA. For high performance face detection, the MCT (Modified Census Transform) method, which is robust against lighting change, was used. The Adaboost learning algorithm was used for creating optimized learning data. And the rotation transformation method was added to maintain effectiveness against face rotating changes. The proposed hardware structure was composed of Color Space Converter, Noise Filter, Memory Controller Interface, Image Rotator, Image Scaler, MCT(Modified Census Transform), Candidate Detector / Confidence Mapper, Position Resizer, Data Grouper, Overlay Processor / Color Overlay Processor. The face detection engine was tested using a Virtex5 LX330 FPGA board, a QVGA grade CMOS camera, and an LCD Display. It was verified that the engine demonstrated excellent performance in diverse real life environments and in a face detection standard database. As a result, a high performance real time face detection engine that can conduct real time processing at speeds of at least 60 frames per second, which is effective against lighting changes and face rotating changes and can detect 32 faces in diverse sizes simultaneously, was developed.

• Journal of Korea Port Economic Association
• /
• v.34 no.4
• /
• pp.53-68
• /
• 2018

This paper analyzes comparatively business performance indicators and determinants of small and medium sized shipping logistics companies in Korea, using 2015 economic census data. For this purpose, this study estimates various business performance indicators according to 2015 small and medium sized companies classification standards, including operating income to sales and gross value-added to sales. In addition, this study analyzes determinants of business performance using generalized least squares models. The results indicate that average sales, operating income and value-added, sales and operating income per worker, operating income to sales, and material cost to sales of large sized companies are higher than those of small and medium sized companies. The business performance indicators differ by industry and size. Moreover, the determinants of business performance are analyzed in terms of the unemployment rate (-), number of employees (-), sales (+), labor cost ratio (+), and labor cost per employee (-) and the impacts of the individual explanatory variables based on elasticity are different. Finally, this quantitative information could be used to improve the business performance of domestic shipping logistics companies.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.21-29
• /
• 2023

In this paper, we propose an AI-based system for automatically classifying industry and occupation codes in the population census. The accurate classification of industry and occupation codes is crucial for informing policy decisions, allocating resources, and conducting research. However, this task has traditionally been performed by human coders, which is time-consuming, resource-intensive, and prone to errors. Our system represents a significant improvement over the existing rule-based system used by the statistics agency, which relies on user-entered data for code classification. In this paper, we trained and evaluated several models, and developed an ensemble model that achieved an 86.76% match accuracy in industry and 81.84% in occupation, outperforming the best individual model. Additionally, we propose process improvement work based on the classification probability results of the model. Our proposed method utilizes an ensemble model that combines transfer learning techniques with pre-trained models. In this paper, we demonstrate the potential for AI-based systems to improve the accuracy and efficiency of population census data classification. By automating this process with AI, we can achieve more accurate and consistent results while reducing the workload on agency staff.

• International Journal of Advanced Culture Technology
• /
• v.10 no.3
• /
• pp.63-70
• /
• 2022

This study is to investigate that population change as a result of the decline in population has a correlation with a decrease in crime, with the change in the demographic composition by comparing with two models: model with growth in population and one with the decline in population. We collected demographic data for all cities in Korea from the 2010 Census to 2020 offered by the Korean Statistical Information Service, with crime data comprising serious reported crime events from the Korean Nation Police Agency through requesting data related to the total number of crimes at the same as the period of demographic data. This study can identify the impacts of demographic changes as a result of population change on crime change through a comparative analysis between areas with population growth and ones with population decline. We can confirm that there are differences in determinants of crime between areas with population increase and one with population decrease from the analysis of the impact of demographic change as a result of population change on crime change.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.1
• /
• pp.91-96
• /
• 2015

In this study, we introduce robust face recognition system with illumination variation realized with the aid of CT preprocessing method. As preprocessing algorithm, Census Transform(CT) algorithm is used to extract locally facial features under unilluminated condition. The dimension reduction of the preprocessed data is carried out by using $(2D)^2$PCA which is the extended type of PCA. Feature data extracted through dimension algorithm is used as the inputs of proposed radial basis function neural networks. The hidden layer of the radial basis function neural networks(RBFNN) is built up by fuzzy c-means(FCM) clustering algorithm and the connection weights of the networks are described as the coefficients of linear polynomial function. The essential design parameters (including the number of inputs and fuzzification coefficient) of the proposed networks are optimized by means of artificial bee colony(ABC) algorithm. This study is experimented with both Yale Face database B and CMU PIE database to evaluate the performance of the proposed system.

• Korea journal of population studies
• /
• v.8 no.1
• /
• pp.87-117
• /
• 1985

In Principle, the distriction should be understood between projections and forecasts. When the author or user of a projection is willing to describe it as indicating the most likely population at a give date, then he has made a forecast Population change since 1 960 has been reviewed briefly in order to forecast the population of Korea in the year 2,000 which is a leading factor in long term national development plan for which Korea Institute for Population and Health (KIPH) has been participated since 1983. The author of this paper introduced the population forecast prepared for the long term national development plan and an attempt of comparisons with other forecasts such as D.P. Smith's, T. Frejka's, Economic Planning Board's (EPB), UN's and S.B. Lee's was made. Those six forecasts of Korean future population in year 2,000 varried from 48.5 million to 50.0 million due to the base population and assumption of fertility and mortality however the range of total population size is not large enough. Taking four forecasts such as KIPH, EPB, UN, and Lee based on 1980 population census results and latest data of fertility and mortality, KIPH and UN forecast are close in total population size even though there was a slight difference in fertility and mortality assumptions. The smallest size of total population was shown by S.B. Lee (see Table 13) although the difference between KIPH and Lee was approximately one million which is two percent of total population in year 2,000. As a summary of conclusion the author pointed out that one can take anyone of forecasts prepared by different body because size and proportion wise of the Korean population until early I 990s can not be different much and new population projections must be provided by using 1985 population census data and other latest fertility and mortality information coflected by Korea Institute for Population and Health and Economic Planning Board in forth comming year.