• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2004.03a
• /
• pp.367-372
• /
• 2004

비교표준지를 이용하여 개별공시지가를 산정하는 우리나라 제도 하에서 가장 중요한 문제는 개별필지 주변의 표준지 중에서 어떤 표준지를 선택ㆍ이용하여 지가를 산정해야 하는가이다. 그러나 지침상에서는 비교표준지 선정시 중요 요인으로 작용하고 있는 유사가격권에 대하여 수치적인 기준이 아닌 모호한 개념상으로 규정하고 있어 비교표준지 선정에 있어 부정확성을 초래하고 있다. 따라서, 본 연구에서는 객관적인 기준과 수치적인 기준의 부재로 많은 문제점을 발생시키고 있는 유사가격권 설정의 문제를 해결하고자 K-평균 군집화 기법을 활용하여 가격권을 설정하고 이에 대한 타당성을 제시하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.853-856
• /
• 2008

최근 모바일 기기 사용자의 수가 증가함에 따라 모바일 기기 사용자의 위치정보를 관리하기 위한 기법들이 활발히 연구되고 있다. 기존의 모바일 시스템에서 위치정보를 관리하기 위한 방법으로 two-tier 방식과 two-tier 방식을 개선한 구조적 기법이 제시되었다. 구조적 기법에서는 어떻게 위치 데이터베이스를 군집화시키는 것이 매우 중요하다. 왜냐하면 데이터베이스를 군집하는 방법에 따라 업데이트 비용의 차이가 크기 때문이다. 구조적 기법을 위한 이전 연구는 set-cover 알고리즘을 기반한 bottom-up 방식의 시스템 이다. 본 논문에서는 구조적 기법의 데이터베이스 군집화를 위해 K-means clustering 알고리즘을 기반한 top-down 방식의 시스템을 사용하였고, 실험을 통해 본 논문에서 제시된 방식의 시스템이 기존 방식의 시스템보다 데이터베이스 업데이트측면에서 13.67%의 성능이 향상되었음을 보였다.

• Journal of Biomedical Engineering Research
• /
• v.24 no.2
• /
• pp.55-60
• /
• 2003

For quantitative analysis of the cardiac diseases. it is necessary to segment the left-ventricle (LY) in MR (Magnetic Resonance) cardiac images. Snake or active contour model has been used to segment LV boundary. However, the contour of the LV front these models may not converge to the desirable one because the contour may fall into local minimum value due to image artifact inside of the LY Therefore, in this paper, we Propose the Preprocessing method using k-means clustering and merging algorithms that can improve the performance of the active contour model. We verified that our proposed algorithm overcomes local minimum convergence problem by experiment results.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2004.12a
• /
• pp.132-138
• /
• 2004

최근 이동전화, PDA, 텔레매틱스 단말기 등과 같은 모바일 기기에서 공간데이터에 대한 사용이 증가하고 있다. 하지만 모바일 기기의 저장 공간이 늘어났음에도 불구하고 여전히 공간데이터에 대한 요구를 수용하기에는 한계가 있다. 따라서 본 연구에서는 모바일 환경에서 사용 가능한 공간데이터에 대한 손실 압축 기법을 제시하고, 실험을 통한 압축률, 데이터 손실률을 분석하여 연구의 타당성과 적용 가능성을 제시하고자 한다. 세부적으로 압축률과 데이터 손실에 따르는 위치 정확도 관계에서 위치정확도를 높일 수 있는 방향을 모색하여 보았다. 그리고 다양한 군집화 기법 중 연구에 적용 가능한 기법을 선정 이용하였다. 또한 저장 공간뿐만 아닌 연산 성능 측면에서도 열악한 모바일 환경에서 만족할 만한 복원 성능을 보여야 한다. 따라서 압축된 데이터를 복원하는데 소요되는 비용을 최소화할 수 있는 방향이 연구되었다.

• Journal of Broadcast Engineering
• /
• v.18 no.4
• /
• pp.643-646
• /
• 2013

We present an improved AdaBoost algorithm to avoid overfitting phenomenon. AdaBoost is widely known as one of the best solutions for object detection. However, AdaBoost tends to be overfitting when a training dataset has noisy samples. To avoid the overfitting phenomenon of AdaBoost, the proposed method divides positive samples into K clusters using k-means algorithm, and then uses only one cluster to minimize the training error at each iteration of weak learning. Through this, excessive partitions of samples are prevented. Also, noisy samples are excluded for the training of weak learners so that the overfitting phenomenon is effectively reduced. In our experiment, the proposed method shows better classification and generalization ability than conventional boosting algorithms with various real world datasets.

• Korean Journal of Construction Engineering and Management
• /
• v.19 no.2
• /
• pp.61-72
• /
• 2018

Conflicts in public construction projects lead to increase social costs as well as construction costs and schedule delay. Therefore, it is important to evaluate the conflict in construction project and find appropriate solutions based on previous cases. In this research, the conflict factors and criteria for evaluating conflict are derived and 30 cases are evaluated by 11 conflict experts. Using k-means clustering, the cases are clustered by three clusters. The cases were analyzed according to the characteristics of each cluster and labeled as 'NIMBY and harmful facility conflict cluster', 'environmental and pollution conflict cluster', and 'PIMFY and small conflicts'. In the future, when conflict occurs in the public construction projects, the conflict can be evaluated using this clustering and the characteristics of the conflicts can be found. As a result, it will be helpful to mitigate the conflict quickly and effectively by looking for previous cases that are suitable for resolving the conflict through appropriate clusters.

• Composites Research
• /
• v.37 no.1
• /
• pp.46-52
• /
• 2024

The development of transportation methods has improved human transportation convenience and made it possible to expand the travel radius of people with disabilities who have difficulty moving. However, in the case of WAV (wheelchair Accessible Vehicle), the safety that may occur in a vehicle accident is still lower than that of regular passenger seats. In particular, in the case of a rear-end collision that may occur in a defenseless situation, it can cause fatal neck injuries to disabled passengers. Therefore, a more detailed design plan must be reflected in the headrest to be applied to WAV. In this study, a multi-cell headrest was proposed to implement local compression characteristic distribution of the headrest during rear-end collision of WAV. Afterwards, a correlation analysis was performed between the passenger's NIC (Neck Injury Criterion) and impact energy absorption using the data set construction through analysis and the clustering results using k-means clustering. As a result of clustering, it was confirmed that data clusters with similar characteristics were formed, and a correlation analysis between NIC and impact energy absorption through the characteristics of each cluster was performed. As a result of the analysis, it was confirmed that the softer the cell compression characteristics in Mid3 and Mid6, the more impact energy absorption increases, and the harder the cell compression characteristics in Front2, Mid3, and Mid6, the more effective it is in reducing NIC.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.529-532
• /
• 2016

본 논문은 기존의 TextRank 알고리즘에 상호정보량 척도를 결합하여 군집 기반에서 키워드 추출하는 ClusterTextRank 기법을 제안한다. 제안 기법은 k-means 군집화 알고리즘을 이용하여 문서들을 여러 군집으로 나누고, 각 군집에 포함된 단어들을 최소신장트리 그래프로 표현한 후 이에 근거한 군집 정보량을 고려하여 키워드를 추출한다. 제안 기법의 성능을 평가하기 위해 여행 관련 블로그 데이터를 이용하였으며, 제안 기법이 기존 TextRank 알고리즘보다 키워드 추출의 정확도가 약 13% 가량 개선됨을 보인다.

• The Journal of Society for e-Business Studies
• /
• v.27 no.2
• /
• pp.21-36
• /
• 2022

Recently, as COVID-19 has made the e-commerce market expand widely, customers who have different consumption patterns appear in the market. Because companies can obtain opinions and information of customers from reviews, they increasingly face the requirements of managing customer reviews on online platform. In this study, we analyze customers and carry out market segmentation for classifying and defining type of customers in e-commerce. Specifically, K-means clustering was conducted on customer review data collected from Wemakeprice online shopping platform, which leads to the result that six clusters were derived. Finally, we define the characteristics of each cluster and propose a customer management plan. This paper is possible to be used as materials which identify types of customers and it can reduce the cost of customer management and make a profit for online platforms.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.8-19
• /
• 2004

In this paper, we propose an efficient method for locating destination address block on both of machine-Printed and handwritten Korean mail piece images. The proposed method extracts connected components from the binary mail piece image, generates text lines by merging them, and then groups the text fines into nine clusters. The destination address block is determined by selecting some clusters. Considering the geometric characteristics of address information on Korean mail piece, we split a mail piece image into nine areas with an equal size. The nine clusters are initialized with the center coordinate of each area. A modified Manhattan distance function is used to compute the distance between text lines and clusters. We modified the distance function on which the aspect ratio of mail piece could be reflected. The experiment done with live Korean mail piece images has demonstrated the superiority of the Proposed method. The success rate for 1, 988 testing images was about 93.56%.