• 응용통계연구
• /
• 제22권4호
• /
• pp.677-686
• /
• 2009

표본조사에서 널리 쓰이는 모집단의 층화는 추정의 효율을 높이는 방법 중의 하나지만, 이상점을 포함하는 변수가 있는 경우에 여러 가지 문제점을 유발시킬 수 있다. 특히, 이상점이 존재하는 다변량 자료의 경우, 층화를 위한 $\kappa$-평균 군집방법은 이상점에 매우 민감하여 추정의 효율을 떨어뜨릴 수 있다. 본 연구에서는 이상점이 존재하는 다변량 자료의 층화를 위해 $\kappa$-평균 군집방법보다 강건하며 이상점을 따로 식별하는 과정이 배제된 $\kappa$-공간중위수 군집방법을 제안한다. 기존 관련연구인 박진우와 윤석훈 (2008)과 동일한 자료에 대한 사례분석을 통해 층화과정들을 비교, 검토하였으며 이들의 효율성을 추정량의 분산을 통해 비교하였다.

• 한국지능시스템학회논문지
• /
• 제14권5호
• /
• pp.517-524
• /
• 2004

1960년대 퍼지 이론이 소개된 이후 데이터 마이닝을 포함한 기계 학습 분야의 군집화 작업에서 퍼지 이론이 폭넓게 사용되었다. 퍼지 C-평균 알고리즘은 가장 많이 사용되는 퍼지 군집화 알고리즘이다. 이 알고리즘은 하나의 데이터 개체가 서로 다른 소속 정도를 가지고 각 군집에 할당될 수 있도록 한다. 퍼지 C-평균 알고리즘도 K-평균 알고리즘과 같은 일반적인 군집화 알고리즘과 마찬가지로 초기 군집수와 군집 중심의 위치에 의해 최종 군집 결과의 성능 차이가 나타난다. 군집화를 위한 이러한 초기 설정은 주관적이며 이 때문에 적절치 못한 결과를 얻게 될 수도 있다. 본 논문에서는 이 문제를 해결할 수 있는 방법으로 주어진 학습 데이터의 속성을 기반으로 한 초기 군집수와 군집 중심을 결정하는 개선된 밀도 기반의 퍼지 C-평균 알고리즘을 제안하였다. 제안 방법은 격자를 사용하여 초기 군집 중심의 위치와 군집수를 결정하였다. 기존에 많이 이용되었던 객관적인 기계 학습 데이터를 이용하여 제안 알고리즘의 성능비교를 수행하였다.

• The Journal of the Acoustical Society of Korea
• /
• 제20권4E호
• /
• pp.31-38
• /
• 2001

A large-scale synthesis database for a unit selection based synthesis method usually retains redundant synthesis unit instances, which are useless to the synthetic speech quality. In this paper, to eliminate those instances from the synthesis database, we proposed a new pruning method called weighted vector quantization (WVQ). The WVQ reflects relative importance of each synthesis unit instance when clustering the similar instances using vector quantization (VQ) technique. The proposed method was compared with two conventional pruning methods through the objective and subjective evaluations of the synthetic speech quality: one to simply limit maximum number of instance, and the other based on normal VQ-based clustering. The proposed method showed the best performance under 50% reduction rates. Over 50% of reduction rates, the synthetic speech quality is not seriously but perceptibly degraded. Using the proposed method, the synthesis database can be efficiently reduced without serious degradation of the synthetic speech quality.

• Journal of Information Processing Systems
• /
• 제20권3호
• /
• pp.296-306
• /
• 2024

Traditional ecological driving (Eco-Driving) evaluations often rely on mathematical models that predominantly offer subjective insights, which limits their application in real-world scenarios. This study develops a robust, data-driven Eco-Driving evaluation model by integrating dynamic and distributed multi-source data, including vehicle performance, road conditions, and the driving environment. The model employs a combination weighting method alongside K-means clustering to facilitate a nuanced comparative analysis of Eco-Driving behaviors across vehicles with identical energy consumption profiles. Extensive data validation confirms that the proposed model is capable of assessing Eco-Driving practices across diverse vehicles, roads, and environmental conditions, thereby ensuring more objective, comprehensive, and equitable results.

• Journal of Information Processing Systems
• /
• 제14권3호
• /
• pp.694-708
• /
• 2018

China possesses a passenger dedicated line system of large scale, passenger flow intensity with uneven distribution, and passenger nodes with complicated relations. Consequently, the significance of passenger nodes shall be considered and the dissimilarity of passenger nodes shall be analyzed in compiling passenger train operation and conducting transportation allocation. For this purpose, the passenger nodes need to be hierarchically divided. Targeting at problems such as hierarchical dividing process vulnerable to subjective factors and local optimum in the current research, we propose a clustering approach based on self-organizing map (SOM) and k-means, and then, harnessing the new approach, hierarchical dividing of passenger dedicated line passenger nodes is effectuated. Specifically, objective passenger nodes parameters are selected and SOM is used to give a preliminary passenger nodes clustering firstly; secondly, Davies-Bouldin index is used to determine the number of clusters of the passenger nodes; and thirdly, k-means is used to conduct accurate clustering, thus getting the hierarchical dividing of passenger nodes. Through example analysis, the feasibility and rationality of the algorithm was proved.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 신호처리소사이어티 추계학술대회 논문집
• /
• pp.369-372
• /
• 2003

In this paper, we propose impr oved binary tree vector quantization based on spatial sensitivity which is one of the human visual properties. We combine the weights based on spatial masking effect according to changes of three primary colors in blocks of images with the process of splitting nodes using eigenvector in binary tree vector quantization. The test results show that the proposed method generates the quantized images with fine color and performs better than the conventional method in terms of clustering the similar regions. Also the proposed method can get the better result in subjective qualify test and PSNR.

• 터널과지하공간
• /
• 제13권4호
• /
• pp.294-303
• /
• 2003

터널이나 사면과 같은 암반 구조물의 거동은 불연속면에 의해 지배적인 영향을 받는다. 따라서 암반내 존재하는 불연속면 자료의 조사 및 분석은 암반구조물 설계 및 시공에 있어서 상당히 큰 중요성을 가진다. 이러한 불연속면의 조사 및 분석 작업 중에서 반드시 거처야 할 작업 중 하나가 절리군을 분별하는 것이다. 기존의 절리군 분석 작업은 대부분 시각적인 방법으로 행해지고 있다. 이 경우 분석자의 주관에 따라 차이를 보일 수 있으며 절리의 방향 정보 외의 다른 추가적인 정보들은 사용하기 힘든 단점이 있다. 본 연구에서는 절리군 분석을 돕기 위하여 퍼지 클러스터링 기법을 이용한 프로그램을 개발하였으며 이를 두 가지 형태의 절리 자료에 대한 절리군 분석에 적용하였다. 적용 결과 퍼지 클러스터링 기법이 다수의 절리 자료에 대한 절리군 분석, 평균방향 및 밀집도 추정에 효과적이며 타당한 방법임을 확인하였다.

• Geomechanics and Engineering
• /
• 제11권4호
• /
• pp.493-513
• /
• 2016

A modified grey clustering method is presented to systematically evaluate the risk of water inrush in karst tunnels. Based on the center triangle whitenization weight function and upper and lower limit measure whitenization weight function, the modified grey evaluation model doesn't have the crossing properties of grey cluster and meets the standard well. By adsorbing and integrating the previous research results, seven influence factors are selected as evaluation indexes. A couple of evaluation indexes are modified and quantitatively graded according to four risk grades through expert evaluation method. The weights of evaluation indexes are rationally distributed by the comprehensive assignment method. It is integrated by the subjective factors and the objective factors. Subjective weight is given based on analytical hierarchy process, and objective weight obtained from simple dependent function. The modified grey evaluation model is validated by Jigongling Tunnel. Finally, the water inrush risk of Shangjiawan Tunnel is evaluated by using the established model, and the evaluation result obtained from the proposed method is agrees well with practical situation. This risk assessment methodology provides a powerful tool with which planners and engineers can systematically assess the risk of water inrush in karst tunnels.

• 대한한의학회지
• /
• 제33권3호
• /
• pp.1-9
• /
• 2012

Objectives: This research was aimed 1) to find out various subjective symptoms cancer patients had, 2) to find out the quantitative difference of symptoms occurring according to the area of cancer, current treatment methods, and relapse of the cancer and 3) to analyze the relationship between the occurrence of symptoms. Methods: This study was a survey about the various subjective symptoms cancer patients had. The subjects were 48 cancer patients in a double-blinded randomized controlled trial about the effectiveness of Korean Red Ginseng (KRG) treatments. Surveys were examined before Korean Red Ginseng (KRG) treatments. The questionnaire consisted of demographic research, patient history and accompanying symptoms. The symptoms part consisted of 4 questions on general symptoms (vertigo, hot flush, rash, insomnia), 2 questions on digestion symptoms (anorexia, dyspepsia), 2 questions on excretions (diarrhea, constipation), and 5 questions on symptoms in different body parts (headache, dryness of lips, chest pain, aphthous ulcer, nose bleeding). The questionnaire was used to research presence of symptoms and relationships between the occurrence of symptoms. Results: A total of 48 patients took part in this survey. Dizziness was the most frequent subjective symptom patients had (64.58%), followed by dry mouth (43.75%), headache (43.75%), insomnia (41.67%), anorexia (37.5%) and chest discomfort (37.5%). The presence of these subjective symptoms was partially related to the patient's history details - type of cancer, method of cancer treatment (chemotherapy, radiotherapy), termination of treatment, presence of pain, fatigue and recurrence, etc. Factor analysis was conducted to analyze the relationship between the occurrence of symptoms. It suggested 5 factors as a result, but there was a limitation that only a low level of correlation was shown among them. Conclusions: We were able to analyze the occurrences and the relationships among them for accompanying subjective symptoms in cancer patients. However the results of the study are limited in that only 48 patients participated. The symptoms in cancer patients showed a tendency for clustering, occurring simultaneously or continuously rather than occurring alone. The study of these symptom clusters is worth further study, for it is similar to the Symptom Differentiation System in Traditional Korean Medicine. For effective approach to the treatment of cancer in Traditional Korean Medicine, further research on the Korean traditional view of cancer patients should be done based on this research, regarding various symptoms and those relationships on a larger scale.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권3호
• /
• pp.1040-1057
• /
• 2014

In this paper, a new bit rate estimation scheme is proposed to determine the bit rate for each subclass in an MPEG-2 TS to H.264/AVC transcoder after dividing an input MPEG-2 TS sequence into several subclasses. Video format transcoding in conventional IPTV and Smart TV services is a time-consuming process since the input sequence should be fully transcoded several times with different bit-rates to decide the bit-rate suitable for a service. The proposed scheme can automatically decide the bit-rate for the transcoded video sequence in those services which can be stored on a video streaming server as small as possible without losing any subject quality loss. In the proposed scheme, an input sequence to the transcoder is sub-classified by hierarchical clustering using a parameter value extracted from each frame. The candidate frames of each subclass are used to estimate the bit rate using a statistical analysis and a mathematical model. Experimental results show that the proposed scheme reduces the bit rate by, on an average approximately 52% in low-complexity video and 6% in high-complexity video with negligible degradation in subjective quality.