• Proceedings of the Korean Statistical Society Conference
• /
• 2004.11a
• /
• pp.171-174
• /
• 2004

기업들이 심화된 경쟁체제 속에서 고객에 대한 보다 심층적인 이해를 필요로 하고 정보기술의 발달로 각 요소활동내용의 데이터화가 가능해짐에 따라 CRM으로 대변되는 고객 정보의 전략적 활용이 매우 중요하게 되었다. 이를 위해 기업은 고객에 대한 이해를 바탕으로 고객관리 및 마케팅을 수행하기 위한 필수적인 도구인 고객세분화를 수행하고 있다. 본 연구에서는 신용카드고객의 카드사용행태에 근거하여 서로 유사한 사용행태를 보이는 고객군으로 세분화하는 과정을 소개한다. 고객이 실제로 카드를 사용하면서 발생시킨 거래정보에만 의존하여 고객세분화를 수행하였으며 이는 마케팅의 관점에서 상당히 의미 있는 내용이라 볼 수 있다. 고객세분화를 위하여 데이터마이닝기법인 k-평균군집방법과 최장연결법에 의한 계보적 군집방법을 활용하였다

• The Korean Journal of Applied Statistics
• /
• v.18 no.3
• /
• pp.555-565
• /
• 2005

To Know customers is very important for the company to survive in its cut-throat competition among coimpetitors. Companies need to manage the relationship with each ana every customer, ant make each of customers as profitable as possible. CRM (Customer relationship management) has emerged as a key solution for managing the profitable relationship. In order to achieve successful CRM customer segmentation is a essential component. Clustering as a data mining technique is very useful to build data-driven segmentation. This paper is concerned with building proper customer segmentation with introducing a credit card company case. Customer segmentation was built based only on transaction data which cattle from customer's activities. Two-step clustering approach which consists of k-means clustering and agglomerative clustering was applied for building a customer segmentation.

• Journal of Digital Contents Society
• /
• v.18 no.8
• /
• pp.1635-1640
• /
• 2017

In this paper, a hierarchical image tree model for weather classification is defined in a single outdoor image, and a weather classification algorithm using image intensity and k-mean segmentation image is proposed. In the first level of the hierarchical image tree model, the indoor and outdoor images are distinguished. Whether the outdoor image is daytime, night, or sunrise/sunset image is judged using the intensity and the k-means segmentation image at the second level. In the last level, if it is classified as daytime image at the second level, it is finally estimated whether it is sunny or foggy image based on edge map and fog rate. Some experiments are conducted so as to verify the weather classification, and as a result, the proposed method shows that weather features are effectively detected in a given image.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2012.07a
• /
• pp.315-317
• /
• 2012

본 논문에서는 인체의 골격 위치와 깊이 정보를 사용하여 주위 환경에 강건한 특성을 지니는 움직이는 사람 영역 검출 방법을 제안한다. 먼저 영상 내에서 인체의 골격 위치를 검출한 다음 인체 골격의 중심이 될 수 있는 지점에 대해 인체의 평균적 깊이 범위 내에서 깊이 세분화를 수행한다. 그리고 깊이 세분화를 통하여 검출된 사람 영역의 후보군에 대해 윤곽선 기반의 움직임 검출기법을 사용하여 후보군 내에서 움직이는 사람에 해당하는 특징점을 검출한다. 마지막으로 잡음 제거 및 움직이는 사람에 해당하는 영역 검출을 위하여 개선된 깊이 세분화 과정을 수행한다.

• Korean Journal of Soil Science and Fertilizer
• /
• v.42 no.2
• /
• pp.88-97
• /
• 2009

This study was conducted to obtain the basic information to increase the practical use of soil survey data through the subdividing of lava shapes with soil sequences due to different elevations in Jeju. The numbers of soil series of lava topography had occupied many of whole soil series in Jeju. When its topography subdivide, it give more detailed soil information. The obtained results are as follows; The lava topography to subdivide lava topography were studied with 38 soil series according to elevation in Jeju. Division of elevation are less than 50m, 50m to 200m, and 200m to 400m and more than 400m. Name the depending on elevation, less than 50m is called lower part of lava, 50m to 200m is called middle part of lava, and 200m to 400m and more than 400m are called upper part of lava. The characteristics of lava subdivide are as follows; soil family texture of lower part of lava is fine silty to clayey, drainage classes are various, average of available soil depth is 75.3cm, average of gravely contents are 11.6%, average of slopeness is 7.2%, limiting factor are various and soil order are various. soil family texture of middle part of lava is fine silty to coarse silty, drainage classes are well to very well, average of available soil depth is 65.9cm, average of gravely contents are 14.7%, average of slopeness is 11.3%, limiting factor are ashy and soil order are Andisols and Inceptisols. Soil family texture of upper part of lave is fine silty, drainage classes are well, average of available soil depth is 72.8cm, average of gravely contents are 16.0%, average of slopeness is 14.9%, limiting factor are ashy and skeletal, and order are Andisols.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.267-269
• /
• 2015

In the modern society, the images provide the most effective information in multi-devices and the edge includes important feature information in such images. This edge is used as an essential preconditioning process in several application fields and many studies have been carried out in order to obtain the excellent images. The methods of Sobel and Roberts which are generally known are simple to implement the images but bring more or less insufficient processing result. Thus, this paper proposed an edge detection algorithm using the area averaging of segmented mask in order to supplement the problems of the current methods and compared it with such current methods.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.50 no.spc1
• /
• pp.94-98
• /
• 2005

This study was carried out to provide the geographic information based on the distribution of rice qualities and palatability in Gyeoungbuk province of Korea. The rice grain quality and environmental factors were analyzed using 513 sampling sites based on different five-agroclimtic regions of Gyeoungbuk province during three years from 2002 to 2004. In rice grain quality characteristics, the average palatability was low in South eastern coastal and Tabaek semi alpine regions as $67.6\~68.3$ and the coefficient of variation (CV) was relatively high as $6.2\~7.4\%$. The average head rice rates were low in South and Central eastern coastal regions as $87.3\~88.2\%$ and CV was high as $8.2\~8.3\%$. The average protein content was high in Central eastern coastal regions as $8.0\%$ and CV was high as $8.2\~8.3\%$. In case of palatability, the variation was differed clearly between high and low agroclimatic regions; it means that it is possible to divide the same agroclimatic region of high CV into two or three areas by CV of palatability. As the results of subdividing each existing agroclimatic regions based on the palatability, the variation of grain quality characteristics was become lower than that of existing five-agroclimatic regions. Therefore, the re-establishing of agroclimatic region based on rice grain quality was very important for precise cultivation for rice.

• The Korean Journal of Applied Statistics
• /
• v.17 no.1
• /
• pp.135-144
• /
• 2004

We propose a reproducibility (validity) assessment procedure of K-means cluster analysis by randomly partitioning the data set into three parts, of which two subsets are used for developing clustering rules and one subset for testing consistency of clustering rules. Also, as an alternative to Rand index and corrected Rand index, we propose an entropy-based consistency measure between two clustering rules, and apply it to determination of the number of clusters in K-means clustering.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.454-456
• /
• 2005

일반적으로 중국어의 명사구는 최단명사구, 기본명사구 최장명사구로 분류된다. 최장명사구에 대한 정확한 식별은 문장의 전체적인 구조를 파악하고 문장의 정확한 지배용언을 찾아내는데 중요한 역할을 한다. 본 논문에서는 특성에 따라 5개의 클래스로 세분화된 문장부호를 학습자질로 사용하여 최장명사구 자동식별을 진행한다. 제안된 기법은 평균길이가 4인 최장명사구의 식별실험에서 기본모델(baseline)보다 $4.5\%$ 향상된 평균 $85.1\%$의 우수한 F-measure 성능을 보인다.

• Annual Conference on Human and Language Technology
• /
• 2005.10a
• /
• pp.112-119
• /
• 2005

명사구는 기본명사구와 최장명사구로 분류된다. 최장명사구에 대한 정확한 식별은 문장의 전체적인 구문구조를 파악하고 문장의 정확한 지배용언을 찾아내는데 중요한 역할을 수행한다. 본 논문에서는 확장된 청크(chunk) 개념과 다섯 개의 클래스로 세분화된 문장부호 정보를 사용한 최장명사구 식별 기법을 제안한다. 제안된 기법은 기본모델(baseline)보다 4.05% 향상된 평균 88.63%의 우수한 F-measure 성능을 보인다.