• 한국의사학회지
• /
• 제28권1호
• /
• pp.93-102
• /
• 2015

Objectives : The aim of this paper is finding the combinations of the medicinal herbs that are used frequently by analyzing the details of the herbal medicinal prescriptions used by the patients who were diagnosed with M54 code diseases. In addition, I will seek to assess the demonstrative pattern that frequently manifest in the M54 code disease patients by using the combinations of the medicinal herbs that are used frequently. Methods : After having extracted the prescription administered to the patients with the diagnostic code of M54, find the relevance with the demonstrative pattern by analyzing the combination for each of the medicinal herb effectiveness. Use the list of medicinal herbs utilized in the corresponding prescription to examine the most frequent combination of the medicinal herbs through the generation of up to 25 arbitrary combinations of the medicinal herbs. Results & Conclusions : As the results of the analysis of the details of the use of the prescribed herbal medicine packages by the Korean Medicine Hospital of Pusan National University, regarding the back pain of the diagnostic code M54, the prescriptions that corresponded to the kidney deficiency pattern, static blood pattern, wind pattern, dampness pattern, food accumulation pattern, qi depression pattern and phlegm-retained fluid pattern back pain among the back pain classifications under the Dongeuibogam (東醫寶鑑) were used frequently, and, regarding the Nape Pain, prescriptions that corresponded to the pain arising from the wind-dampness and phlegm the 'Taeyang meridian' was most frequent.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권4호
• /
• pp.260-263
• /
• 2008

In this paper we propose Personal Information Management System for Library Database. It manages personal search pattern for the given user and provide specific book list for library book search system. With the proposed system, the conventional overlap searching time will be decreased with personalized information and search history. This system manages the individual data according to personal searching pattern, sequence and usability. Therefore, the user can locate necessary book information more accurately with their distinct interest and search history.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.95-97
• /
• 2004

This paper estimates Adaptive Resonance Theory 1(ART1) as a vigilance parameter of pattern clustering algorithm. Inherent characteristics of the model are analyzed. In particular the vigilance parameter ${\rho}$ and its role in classification of patterns is examined. Our estimates show that the vigilance parameter as designed originally does not necessarily increase the number of categories with its value but can decrease also. This is against the claim of solving the stability-plasticity dilemma. However, we have proposed a modified vigilance parameter estimate criterion which takes into account the problem of subset and superset patterns and stably categorizes arbitrarily many input patterns in one list presentation when the vigilance parameter is closer to one.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제21권1호
• /
• pp.19-28
• /
• 2015

본 논문에서는 과거 교통정체 패턴을 이용하여 현재의 교통정체가 풀리는 정체인지 아니면 악화되는 정체인지를 판별하는 알고리즘을 제안한다. 과거 교통정체 패턴은 다중 포인터를 이용하여 정체구간들을 연결한 인접 리스트에 교통정체의 시간적 길이와 공간적 길이로 저장된다. 교통정체가 시작된 구간에 해당하는 헤드노드를 탐색하고 현재패턴과 가장 유사한 과거 교통정체 패턴을 이용하여 장래의 교통정체 변화정보를 제공한다. 실험을 통해 검증한 결과, 도로 구간 하나에 대한 정체 변화를 판별하였을 때 실제 값과 비교해서 평균적으로 15분 오차를 보였으며, 연속된 다수의 도로 구간들을 결합하여 비교적 긴 구간의 정체 변화를 판별하였을 경우 평균적으로 10분 이내의 오차를 보이며 실제 값과 유사한 것을 보였다.

• 한국정보통신학회논문지
• /
• 제12권7호
• /
• pp.1243-1248
• /
• 2008

최근 생명 정보학 기술의 발달로 마이크로 단위의 실험조작이 가능해짐에 따라 하나의 chip상에서 전체 genome의 expression pattern을 관찰할 수 있게 되었고, 동시에 수 만개의 유전자들 간치 상호작용도 연구 가능하게 되었다. 본 논문에서는 암에 걸린 흰쥐 외피 기간 세포 분화 실험에서 얻어진 3840 유전자의 마이크로어레이 cDNA를 이용해 데이터의 정규화를 거쳐 본 논문에서 제안한 유사성 척도 조합 방법으로 정보력 있는 유전자들을 추출한 후, 유사성 척도 조합 방법과 결합한 멀티퍼셉트론 신경망 분류기와 기존의 DT, NB, SVM 분류기를 이용하여 클래스 분류 시스템을 구축하고, 성능을 비교분석하였다. 피어슨 적률 상관 계수와 유클리디안 거리 계수 조합을 이용하여 선택된 200 유전사들을 멀티퍼셉트론 신경망 분류기로 분류한 결과 98.84%의 정확도를 보여 다른 분류기를 이용하여 실험을 수행한 경우보다 향상된 분류 성능을 보였다.

• 대한예방한의학회지
• /
• 제19권3호
• /
• pp.11-19
• /
• 2015

Objective : This study aimed to develop and preliminary validate the Mibyeong Questionnaire (MQ) based on Deficiency-Stagnation pattern, which was report of a person's health condition that comes directly from the person. Method : The first phage of developing a MQ was to generate and exhaustive list of all MQ issue that are relevant to the domain of interest, using literature research and expert group discussions. Through those steps, we established MQ with 33 items divided into two parts: 21-item deficiency pattern and 12-item stagnation pattern. The second phage, we examined the preliminary tests of reliability and validity including the 16-item Deficiency of MQ (16D-MQ), with data (n=1,890) already collected on the Korean medicine data center in KIOM. Results : Exploratory factor analysis revealed three factors of the 16D-MQ. These factors were fatigue(Qi, 氣); psychic and physical elements(Shen, 神); and skin and hair(Jing, 精). Cronbach's coefficient alpha was 0.876 and the intraclass correlation coefficients was 0.368-0.538. In support of criteria validity, the 16D-MQ was weakly correlated with EQ-5D and physicians's opinion, but it was acceptable. Conclusion : The MQ shows that it has an appropriate level of internal consistency and validity. We think further study to reveal its reliability and validity, including stagnation pattern as well as deficiency pattern, is needed.

• 인터넷정보학회논문지
• /
• 제16권6호
• /
• pp.23-30
• /
• 2015

빅 데이터가 이슈화됨에 따라 데이터 분석의 결과를 기반으로 동작하는 많은 응용들이연구되고 왔고, 대표적인 응용들은 전자상거래 시스템의 상품 추천 서비스, 검색 엔진에서의 검색 서비스, 소셜 네트워크 서비스에서의 친구 추천 서비스 등이 있다. 본 논문은 기존의 데이터 마이닝 기법 중 데이터 집합에서 나타나는 유사한 패턴들을 마이닝하는 빈발 패턴 트리와 컴퓨터 과학의 이론에 기초한 결정트리를 결합하여 결정 빈발 트리 알고리즘을 제안한다. 이는 기존의 빈발 패턴 트리 알고리즘은 패튼 트리에서 패턴 생성에 대한 정확성은 보장되나 소셜 데이터처럼 다양한 패턴이 나타는 데이터에 대해서는 많은 수의 패턴들을 생성시켜 분석에 대한 어려움이 있어, 서브트리들과의 수렴 여부를 판단하는 모델로 변형시켜 문제를 개선한다. 또한 맵리듀스로 모델링하여 분산처리를 통한 고속 처리 알고리즘을 제시한다.

• 복식
• /
• 제67권1호
• /
• pp.56-73
• /
• 2017

The purpose of this study is to categorize the lotus patterns of traditional Korean fabrics. This study collected 169 fabrics of Korean lotus flower patterns and then made a list of era, configuration of pattern, type of artifact, findspot and holding institutions. It analyzes the characteristics of 251 kinds of lotus woven on 169 Korean fabrics, 251 kinds of lotus on 169 fabrics divided into Realistic Type, Design Type, and Abstract Type. They consist of 213 kinds of Design Type(84.8%), 21 kinds of Realistic Type(8.4%), and 17 kinds of Abstract Type(6.8%). The largest part of lotus patterns is Design Type. Design Types are subdivided into 14 types. This result contrasted with the conclusion of the research paper about peony patterns in Traditional Korean Textiles. The largest part of peony patterns was Realistic Types. Realistic Types of lotus patterns in Korean textiles are subdivided into 6 types. Korea has more diverse forms and a higher ratio of Realistic Types than China and this can be interpreted to reflect that there is a tendency in Korea to prefer natural patterns. Besides abstract Types are subdivided into 3 types. When the lotus flower patterns of Korea and China are categorized and similar types are compared, there are some unique patterns that appear only in Korea or China. Not only that, similar types display differences in pattern arrangement methods and portrayal among the two countries. If such study results are utilized, they can be grounds for distinguishing the production area of lotus flower textile fabrics when they are newly excavated in the future.

• 동의생리병리학회지
• /
• 제22권2호
• /
• pp.453-458
• /
• 2008

This study aimed to develop an evaluation reliability of Korean pattern identification for wind stroke. We studied 643 patients with stroke and made a list of registry for each of them. The present study analyzed 553 cases, in which the resident’s pattern identification agreed with the specialist’s one, and the cases included five differentiation pattern: the fire-heat pattern (114), the dampness-phlegm pattern (157), the static blood pattern (11), the Yin deficiency pattern (81), and the Qi deficiency pattern (190). This study showed that none of the Cronbach's alpha reached 0.700, which is the general reliable level. The average Cronbach's alpha of each symptoms was 0.353 for the dampness-phlegm pattern, 0.571 for the fire-heat pattern, 0.443 for the Qi deficiency pattern, 0.451 for the Yin deficiency pattern, and 0.302 for the static blood pattern. This suggests the possibility that each pattern identification could be coincided with other symptoms, and it also shows the limits of pattern identification of this study that narrows the symptoms of paralysis patients into only a single pattern. Continuous compliments and researches should be done referring to this matter. However, the internal consistency analysis of all the pattern identification showed that every Cronbach's alpha were within the range of 0.670 to 0.703, and the Cronbach's alpha of the whole symptoms was evaluated as 0.692, which makes the reliability of the pattern identification as itself almost satisfactory to the general reliable level, and therefore, significant. In the future, continuous clinical research to develope this pattern identification for wind stroke actually applicable to stroke patients needs to be made through accumulating more cases, improving the objectivity.

• 한국도로학회논문집
• /
• 제17권4호
• /
• pp.1-9
• /
• 2015

PURPOSES : The purpose of this study is to revise the weight to power ratio of the representative truck in S. Korea. So far, S. Korea has been using the unit lb/hp, and the construction machines were not considered in the evaluation of the performance of trucks. METHODS : This study was performed to recommend the use of ISO system of units, and to analyze the weight to power ratios of the representative trucks in S. Korea, including the dump trucks, concrete mixer trucks, and asphalt and concrete diffusers. RESULTS: From this study, the 85 percentile value of the weight to power ratio of the trucks in S. Korea's was found to be 103.6 kg/kw. CONCLUSIONS : The performance standard for the representative truck has to be increased upward, considering the travel pattern of the dump trucks, concrete mixers, and asphalt and concrete diffuser trucks, travel distances, narrow area (work zone) of operation, and the saving in construction budget for climbing lane. Based on this study, the weight to power ratio of the representative truck in S. Korea could possibly be revised to 100~110 kg/kW.