• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.5
• /
• pp.133-148
• /
• 2022

In this study, a road pavement crack deterioration model was developed for a pavement road sections of the Sejong-city. Data required for model development were acquired using a deep learning-based road asset monitoring system. Road pavement monitoring was conducted on the same sections in 2021 and 2022. The developed model was analyzed by dividing it into a method for estimating the annual average amount of deterioration and a method based on Bayesian Markov Mixture Hazard model. As a result of the analysis, it was found that an analysis results similar to the crack deterioration model developed based on the data acquired from the Automatic pavement investigation equipmen was derived. The results of this study are expected to be used as basic data by local governments to establish road management plans.

• Journal of Korea Spatial Information System Society
• /
• v.7 no.2 s.14
• /
• pp.15-27
• /
• 2005

The purpose of this study is to present a methodology of GIS audit system which fully reflects standardization regarding GIS. With the recognition of the problems stemming from data exchange, building costs, and budget waste, this study utilized the standardization for evaluation factors of GIS audit. After analyzing the advantages and disadvantages of current audit methodology, this study pointed out the importance of audit, then presenting main audit factors followed by analyzing national standardization and extracting contents of audit to be added into the existing auditing items. Through the analysis of different types and characteristics, and both advantages and disadvantages of GASP, this study identifies and introduces a highly improved and practical methodology called Modified-GASP (M-GASP) that is basically set to be complementary and supplementary to GASP. Ultimately, the result of this study will support the higher degree of efficiency, stability, and extendability of GIS system, not to mention of strengthening the competitiveness of organizations involved.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.1
• /
• pp.69-88
• /
• 2023

Efficient spatial planning is one of the necessary factors to successfully respond to climate change. And researchers often use LULC(Land-Use/Cover) data to conduct land use and spatial planning research. However, LULC data has a limited number of grades related to urban surface, so each different urban structure appearing in several cities is not easily analyzed with existing land cover products. This limitation of land cover data seems to be overcome through LCZ(Local Climate Zone) data used in the urban heat island field. Therefore, this study aims to first discuss whether LCZ data can be applied not only to urban heat island fields but also to other fields, and secondly, whether LCZ data still have problems with existing LULC data. Research methodology is largely divided into two categories. First, through literature review, studies in the fields of climate, land use, and urban spatial structure related to LCZ are synthesized to analyze what research LCZ data is currently being used, and how it can be applied and utilized in the fields of land use and urban spatial structure. Next, the GIS spatial analysis methodology is used to analyze whether LCZ still has several errors that are found in the LULC.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.137-139
• /
• 2023

MARPOL (Maritime Pollution Treaty) 부속서 (Annex VI) 경우 대기오염 규제 내용이며, 최근 선박 배기가스 배출규제 강화 목적 개정됨에 따라 관련 규정을 충족하기 위한 적극적인 조치가 요구된다. 예를 들어 Sulfur Dioxide (SOx, 황산화물) 경우 2020년 기준 전 세계 해역 내 운항하는 선박은 황 함유량 0.5% 기준의 배출 요구 기준을 충족해야 한다. 배출규제 해역 지정 시 해당 해역 선박 배기가스 배출량 계산 산정 기준 확립이 필요함에 따라 대기환경에 대한 종합적인 분석이 필수적으로 요구된다. 본 연구에서는 황산화물 배출량 산출 방법론으로써 그리드 셀 내 선박 점유시간을 계산한다. 점유시간이 길수록 선박 통항 및 배기가스 배출이 밀집되어 있음을 의미한다. 밀집도 분석에 더불어 선박의 특성이 반영된 시간당 배출량을 적용하여 배출 공간 인벤토리를 구축하였으며 분석 결과를 GIS (Geographical Information System) 환경에서 공간 지도로 시각화하였다. 기존 국내 황산화물 배출규제 해역의 효과 평가에 더불어 외항 범위까지 규제 확대 시나리오확립 후 비교 평가를 통하여 배출량 감소 효과를 정량적으로 확인하였다.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.51-62
• /
• 2022

AI-based speed prediction studies have been conducted quite actively. However, while the importance of explainable AI is emerging, the study of interpreting and reasoning the AI-based speed predictions has not been carried out much. Therefore, in this paper, 'Explainable Deep Graph Neural Network (GNN)' is devised to analyze the speed prediction and assess the nearby road influence for reasoning the critical contributions to a given road situation. The model's output was explained by comparing the differences in output before and after masking the input values of the GNN model. Using TOPIS traffic speed data, we applied our GNN models for the major congested roads in Seoul. We verified our approach through a traffic flow simulation by adjusting the most influential nearby roads' speed and observing the congestion's relief on the road of interest accordingly. This is meaningful in that our approach can be applied to the transportation network and traffic flow can be improved by controlling specific nearby roads based on the inference results.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.330-337
• /
• 2018

Recently, there has been a growing interest in user-centered smart city services. In this study, user behavior analysis was performed as a preliminary study for user - centered smart city service planning. In particular, we will use GIS based location analysis data and video ethonography methodology to derive smart city service direction and needs. In this study, the area of Daejeon Design District selected as the Smart City Test bed was selected as the survey area and the location analysis data of the traffic accident analysis system of the road traffic corporation and the fixed camera We observed user's behavior type and change with image data extracted through the technique. Location analysis data is classified according to the type of accident, and image data is classified into 11 subdivided types of user activities. The problems and specificities observed were analyzed. The user behavior characteristics investigated through this study are meaningful to provide a basis for suggesting user - centered smart city services in the future.

• Knowledge Management Research
• /
• v.22 no.4
• /
• pp.1-26
• /
• 2021

Measuring the intangible asset has been vigorously studied for its importance. Especially, the value of character in media industry is difficult to quantitatively evaluate in spite of the industry's rapid growth. Recently, the Social Network Analysis (i.e., SNA) has been actively applied to understand human usage patterns in a media field. By using SNA methodology, this study attempts to investigate how the character network characteristics of media works are linked to human search behaviors. Our analysis reveals the positive correlation and causality between character network centralities and character search data. This result implies that the character network can be used as a clue for the valuation of character assets.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.43-50
• /
• 2024

This paper evaluates the Korean sentiment analysis performance of large language models like GPT-3.5 and GPT-4 using a zero-shot approach facilitated by the ChatGPT API, comparing them to pre-trained Korean models such as KoBERT. Through experiments utilizing various Korean sentiment analysis datasets in fields like movies, gaming, and shopping, the efficiency of these models is validated. The results reveal that the LMKor-ELECTRA model displayed the highest performance based on F1-score, while GPT-4 particularly achieved high accuracy and F1-scores in movie and shopping datasets. This indicates that large language models can perform effectively in Korean sentiment analysis without prior training on specific datasets, suggesting their potential in zero-shot learning. However, relatively lower performance in some datasets highlights the limitations of the zero-shot based methodology. This study explores the feasibility of using large language models for Korean sentiment analysis, providing significant implications for future research in this area.

• Broadcasting and Media Magazine
• /
• v.25 no.2
• /
• pp.17-26
• /
• 2020

최근 스마트폰에서의 증강현실, 미적 효과의 증대(예, 라이브 포커싱) 등의 어플리케이션을 제공하기 위해 모바일 기기에서의 3차원 공간 복원 기술에 대한 관심이 증가하고 있다. 소비자들의 요구에 발 맞춰 최근 스마트폰 제조사는 모든 플래그십 모델에 다중 카메라 및 뎁스 센서(거리 측정 센서)를 탑재하는 추세이다. 본 고에서는 모바일 폰에 탑재되고 있는 대표적인 세 축의 뎁스 추정(공간 복원) 방식에 대해 간단히 살펴보고, 최근 심층학습(Deep learning)의 등장으로 기술 발전의 새로운 국면에 접어 든 다중 시점 매칭(Multi-view stereo) 방법에 대해 소개하고자 한다. 심층 신경망이 재조명 받은 2012년 전까지 주류 연구 방향이었던 전통 기하학 기반의 방법에 대한 소개를 시작으로 심층 신경망기반의 방법론으로의 발전된 형태를 살펴본다. 또한, 신경망기반의 방법론은 크게 3 세대로 나누어 각 세대별 특징에 대해 자세히 살펴보고, 다양한 데이터에 대한 실험 결과를 통해 세대별 공간 복원 결과를 비교 분석한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06b
• /
• pp.101-106
• /
• 2007

본 논문에서는 SOA(Service Oriented Architecture) 기반으로 국가 R&D 정보의 종합 조회 기능을 제공하는 국가 R&D 정보시스템(RnDIS: R&D Information System)을 설계 및 구현하였다. 물리적으로 분산되고 각각 별도의 DB를 구성하여 활용하는 이질적인 4개의 응용시스템의 기능을 효과적으로 연계 및 활용하기 위해 유연하며 확장이 용이한 SOA를 채택하였다. 서비스의 식별, 정의, 분석 등의 개발을 위해 CBD 방법론을 확장한 새로운 서비스 개발방법론을 정의 및 활용하였으며, RnDIS를 위해 4개의 어플리케이션 서비스와 4개의 비즈니스 프로세스 서비스를 정의 및 설계하였다. 어플리케이션 서비스는 기존의 자바코드로부터 WSDL(Web Service Description Language)을 생성하는 래핑(wrapping) 방식을 사용하여 구현하였며, 비즈니스 프로세스 서비스는 BPEL(Business Process Execution Language) 엔진을 이용하여 어플리케이션 서비스를 조합하는 방식을 이용하여 구현하였다. RnDIS는 NTIS(National Science and Technology Information System) 공식 홈페이지(http://www.ntis.go.kr)의 종합검색 메뉴로 시범서비스 되고 있으며, 향후 서비스 대상 데이터의 확장과 기능 추가를 통해 정식 서비스를 오픈 할 예정이다.