• 제목/요약/키워드: one class classification

검색결과 348건 처리시간 0.035초

Angle씨 분류 I급 부정교합의 임상적 고찰 (CLINICAL CONSIDERATION OF ANGLE'S CLASSIFICATION CLASS I MALOCCLUSION)

  • 강흥구
    • 대한치과의사협회지
    • /
    • 제15권2호
    • /
    • pp.107-110
    • /
    • 1977
  • Class I malocclusion is essentially a dental dysplasia. Rotations, individual tooth malpositions, missing teeth, tooth size discrepancies, etc., fall under this classification. There are two types of class I malocclusions. One is identified by and insufficient denture base to accommodate the teeth; the other has more denture base than tooth material, creating spaces in the arch. The tooth material-to denture base discrepancies may be slight, calling for only a little increase in arch length for alignment and the correction of minor rotations. Discrepancies may also be great, in which case it becomes necessary to reduce tooth material by extraction, so as to make the tooth material more in proportion to the size of the denture base. The author had attempted orthodontic treatment of a class I malocclusion case of 13-year old boy in which high canines and impacted mandibular second premolars were involved. The author obtained good results.

  • PDF

New Splitting Criteria for Classification Trees

  • Lee, Yung-Seop
    • Communications for Statistical Applications and Methods
    • /
    • 제8권3호
    • /
    • pp.885-894
    • /
    • 2001
  • Decision tree methods is the one of data mining techniques. Classification trees are used to predict a class label. When a tree grows, the conventional splitting criteria use the weighted average of the left and the right child nodes for measuring the node impurity. In this paper, new splitting criteria for classification trees are proposed which improve the interpretablity of trees comparing to the conventional methods. The criteria search only for interesting subsets of the data, as opposed to modeling all of the data equally well. As a result, the tree is very unbalanced but extremely interpretable.

  • PDF

Unsupervised one-class classification for condition assessment of bridge cables using Bayesian factor analysis

  • Wang, Xiaoyou;Li, Lingfang;Tian, Wei;Du, Yao;Hou, Rongrong;Xia, Yong
    • Smart Structures and Systems
    • /
    • 제29권1호
    • /
    • pp.41-51
    • /
    • 2022
  • Cables are critical components of cable-stayed bridges. A structural health monitoring system provides real-time cable tension recording for cable health monitoring. However, the measurement data involve multiple sources of variability, i.e., varying environmental and operational factors, which increase the complexity of cable condition monitoring. In this study, a one-class classification method is developed for cable condition assessment using Bayesian factor analysis (FA). The single-peaked vehicle-induced cable tension is assumed to be relevant to vehicle positions and weights. The Bayesian FA is adopted to establish the correlation model between cable tensions and vehicles. Vehicle weights are assumed to be latent variables and the influences of different transverse positions are quantified by coefficient parameters. The Bayesian theorem is employed to estimate the parameters and variables automatically, and the damage index is defined on the basis of the well-trained model. The proposed method is applied to one cable-stayed bridge for cable damage detection. Significant deviations of the damage indices of Cable SJS11 were observed, indicating a damaged condition in 2011. This study develops a novel method to evaluate the health condition of individual cable using the FA in the Bayesian framework. Only vehicle-induced cable tensions are used and there is no need to monitor the vehicles. The entire process, including the data pre-processing, model training and damage index calculation of one cable, takes only 35 s, which is highly efficient.

연합판막질환의 판치환수술 (Double Valve Replacement: report of 5 cases)

  • 노중기
    • Journal of Chest Surgery
    • /
    • 제12권4호
    • /
    • pp.355-360
    • /
    • 1979
  • Mitral and aortic valve replacement with tricuspid annuloplasty was undertaken in 5 patients out of 38 valvular surgery between the period from Jan. 1977 to May 1979 in the Dept. of Thoracic and Cardiovascular Surgery in Korea University Hospital. All were male patients with age ranging from 18 to 42 years, and preoperative evaluation revealed one case in Class IV, and four cases in Class III according to the classification of NYHA. Preoperative diagnosis was confirmed by routine cardiac study including retrograde aorto- and left ventriculography, and there were two cases with MSi+ASi+Ti, two cases with MSi+Ai+Ti, and one case with Mi+Ai+Ti. Double valve replacement was performed under the hypothermic cardiopulmonary bypass with total pump time of 247 min. in average ranging from 206 min. to 268 min. During aortic valve replacement, left coronary perfusion was done in the first two cases, and cardiac arrest with cardioplegic solution proposed by Bretschneider was applied in the remained three cases. Starr-Edwards, Bjork-Shiley prosthetic valves and Carpentier-Edwards tissue valve were replaced in the aortic area, and Carpentier-Edwards and Angell-Shiley tissue valves were replaced in the mitral area with each individual combination [three prosthetic and two tissue valves in the aortic, and five tissue valves in the mitral area respectively]. Postoperative recovery was uneventful in all cases except one case with hemopericardium, which was managed with pericardiectomy on the postoperative 10th day in good result. Follow-up after double valve replacement of the all five cases for the period from 6 months to 33 months revealed satisfactory adaptation in social activity and occupation with cardiac function of Class I according to the classification of NYHA In all five cases.

  • PDF

Classification of Mental States Based on Spatiospectral Patterns of Brain Electrical Activity

  • Hwang, Han-Jeong;Lim, Jeong-Hwan;Im, Chang-Hwan
    • 대한의용생체공학회:의공학회지
    • /
    • 제33권1호
    • /
    • pp.15-24
    • /
    • 2012
  • Classification of human thought is an emerging research field that may allow us to understand human brain functions and further develop advanced brain-computer interface (BCI) systems. In the present study, we introduce a new approach to classify various mental states from noninvasive electrophysiological recordings of human brain activity. We utilized the full spatial and spectral information contained in the electroencephalography (EEG) signals recorded while a subject is performing a specific mental task. For this, the EEG data were converted into a 2D spatiospectral pattern map, of which each element was filled with 1, 0, and -1 reflecting the degrees of event-related synchronization (ERS) and event-related desynchronization (ERD). We evaluated the similarity between a current (input) 2D pattern map and the template pattern maps (database), by taking the inner-product of pattern matrices. Then, the current 2D pattern map was assigned to a class that demonstrated the highest similarity value. For the verification of our approach, eight participants took part in the present study; their EEG data were recorded while they performed four different cognitive imagery tasks. Consistent ERS/ERD patterns were observed more frequently between trials in the same class than those in different classes, indicating that these spatiospectral pattern maps could be used to classify different mental states. The classification accuracy was evaluated for each participant from both the proposed approach and a conventional mental state classification method based on the inter-hemispheric spectral power asymmetry, using the leave-one-out cross-validation (LOOCV). An average accuracy of 68.13% (${\pm}9.64%$) was attained for the proposed method; whereas an average accuracy of 57% (${\pm}5.68%$) was attained for the conventional method (significance was assessed by the one-tail paired $t$-test, $p$ < 0.01), showing that the proposed simple classification approach might be one of the promising methods in discriminating various mental states.

Deep Belief Network를 이용한 뇌파의 음성 상상 모음 분류 (Vowel Classification of Imagined Speech in an Electroencephalogram using the Deep Belief Network)

  • 이태주;심귀보
    • 제어로봇시스템학회논문지
    • /
    • 제21권1호
    • /
    • pp.59-64
    • /
    • 2015
  • In this paper, we found the usefulness of the deep belief network (DBN) in the fields of brain-computer interface (BCI), especially in relation to imagined speech. In recent years, the growth of interest in the BCI field has led to the development of a number of useful applications, such as robot control, game interfaces, exoskeleton limbs, and so on. However, while imagined speech, which could be used for communication or military purpose devices, is one of the most exciting BCI applications, there are some problems in implementing the system. In the previous paper, we already handled some of the issues of imagined speech when using the International Phonetic Alphabet (IPA), although it required complementation for multi class classification problems. In view of this point, this paper could provide a suitable solution for vowel classification for imagined speech. We used the DBN algorithm, which is known as a deep learning algorithm for multi-class vowel classification, and selected four vowel pronunciations:, /a/, /i/, /o/, /u/ from IPA. For the experiment, we obtained the required 32 channel raw electroencephalogram (EEG) data from three male subjects, and electrodes were placed on the scalp of the frontal lobe and both temporal lobes which are related to thinking and verbal function. Eigenvalues of the covariance matrix of the EEG data were used as the feature vector of each vowel. In the analysis, we provided the classification results of the back propagation artificial neural network (BP-ANN) for making a comparison with DBN. As a result, the classification results from the BP-ANN were 52.04%, and the DBN was 87.96%. This means the DBN showed 35.92% better classification results in multi class imagined speech classification. In addition, the DBN spent much less time in whole computation time. In conclusion, the DBN algorithm is efficient in BCI system implementation.

단일 클래스 분류와 특징 선택에 기반한 향상된 이상 감지 (Improved Fault Detection Based on One-Class Classification and Feature Selection)

  • 조현우
    • 한국산학기술학회논문지
    • /
    • 제20권8호
    • /
    • pp.216-223
    • /
    • 2019
  • 생산 공정에서 발생되는 공정 이상을 적시에 감지하는 것은 생산 공정의 안전하고 일관된 조업 및 운영에 필수적인 요소 중 하나로서 반드시 필요하다. 예측되지 못하거나 적절하게 감지되지 못한 공정 이상은 전체 생산 공정과 공정에서 생산되는 최종 제품의 품질에 심각한 영향을 줄 수 있기 때문이다. 또한 이러한 상황은 공정 기능 불량과 고장으로 이어지게 된다. 이러한 공정 이상을 신뢰성 있게 적시에 검출하기 위해 본 연구에서는 새로운 단일 클래스 분류에 기반한 공정 이상 감지 기법을 제안한다. 본 연구의 제안된 방법은 잡음 필터링, 특징 선택, 비선형 표현 및 특이치 검출의 네단계로 구성된다. 본 연구에서는 시뮬레이션 공정의 측정치를 활용하여 제안된 방법의 성능을 평가하였다. 그 결과 제안된 공정 이상 탐지 기법이 신뢰할 수 있는 모니터링 결과를 산출하였으며 기존 비교 대상 방법들보다 평균 25.4% 향상된 성능을 보여 주었다. 또한 적합한 특징 선택을 통하여 보다 향상된 이상 감지 성능을 얻을 수 있었다.

불균형 데이터 분류를 위한 딥러닝 기반 오버샘플링 기법 (A Deep Learning Based Over-Sampling Scheme for Imbalanced Data Classification)

  • 손민재;정승원;황인준
    • 정보처리학회논문지:소프트웨어 및 데이터공학
    • /
    • 제8권7호
    • /
    • pp.311-316
    • /
    • 2019
  • 분류 문제는 주어진 입력 데이터에 대해 해당 데이터의 클래스를 예측하는 문제로, 자주 쓰이는 방법 중의 하나는 주어진 데이터셋을 사용하여 기계학습 알고리즘을 학습시키는 것이다. 이런 경우 분류하고자 하는 클래스에 따른 데이터의 분포가 균일한 데이터셋이 이상적이지만, 불균형한 분포를 가지고 경우 제대로 분류하지 못하는 문제가 발생한다. 이러한 문제를 해결하기 위해 본 논문에서는 Conditional Generative Adversarial Networks(CGAN)을 활용하여 데이터 수의 균형을 맞추는 오버샘플링 기법을 제안한다. CGAN은 Generative Adversarial Networks(GAN)에서 파생된 생성 모델로, 데이터의 특징을 학습하여 실제 데이터와 유사한 데이터를 생성할 수 있다. 따라서 CGAN이 데이터 수가 적은 클래스의 데이터를 학습하고 생성함으로써 불균형한 클래스 비율을 맞추어 줄 수 있으며, 그에 따라 분류 성능을 높일 수 있다. 실제 수집된 데이터를 이용한 실험을 통해 CGAN을 활용한 오버샘플링 기법이 효과가 있음을 보이고 기존 오버샘플링 기법들과 비교하여 기존 기법들보다 우수함을 입증하였다.

일 대학병원 호스피스 병동 입원 환자의 간호활동시간 측정과 원가산정 (Determination of Cost and Measurement of nursing Care Hours for Hospice Patients Hospitalized in one University Hospital)

  • 김경운
    • 간호행정학회지
    • /
    • 제6권3호
    • /
    • pp.389-404
    • /
    • 2000
  • This study was designed to determine the cost and measurement of nursing care hours for hospice patients hostpitalized in one university hospital. 314 inpatients in the hospice unit 11 nursing manpower were enrolled. Study was taken place in C University Hospital from 8th to 28th, Nov, 1999. Researcher and investigator did pilot study for selecting compatible hospice patient classification indicators. After modifying patient classification indicators and nursing care details for general ward, approved of content validity by specialist. Using hospice patient classification indicators and per 5 min continuing observation method, researcher and investigator recorded direct nursing care hours, indirect nursing care hours, and personnel time on hospice nursing care hours, and personnel time on hospice nursing care activities sheet. All of the patients were classified into Class I(mildly ill), Class II (moderately ill), Class III (acutely ill), and Class IV (critically ill) by patient classification system (PCS) which had been carefully developed to be suitable for the Korean hospice ward. And then the elements of the nursing care cost was investigated. Based on the data from an accounting section (Riccolo, 1988), nursing care hours per patient per day in each class and nursing care cost per patient per hour were multiplied. And then the mean of the nursing care cost per patient per day in each class was calculated. Using SAS, The number of patients in class and nursing activities in duty for nursing care hours were calculated the percent, the mean, the standard deviation respectively. According to the ANOVA and the $Scheff{\'{e}$ test, direct nursing care hours per patient per day for the each class were analyzed. The results of this study were summarized as follows : 1. Distribution of patient class : class IN(33.5%) was the largest class the rest were class II(26.1%) class III(22.6%), class I(17.8%). Nursing care requirements of the inpatients in hospice ward were greater than that of the inpatients in general ward. 2. Direct nursing care activities : Measurement ${\cdot}$ observation 41.7%, medication 16.6%, exercise ${\cdot}$ safety 12.5%, education ${\cdot}$ communication 7.2% etc. The mean hours of direct nursing care per patient per day per duty were needed ; 69.3 min for day duty, 64.7 min for evening duty, 88.2 min for night duty, 38.7 min for shift duty. The mean hours of direct nursing care of night duty was longer than that of the other duty. Direct nursing care hours per patient per day in each class were needed ; 3.1 hrs for class I, 3.9 hrs for class II, 4.7 hrs for class III, and 5.2 hrs for class IV. The mean hours of direct nursing care per patient per day without the PCS was 4.1 hours. The mean hours of direct nursing care per patient per day in class was increased significantly according to increasing nursing care requirements of the inpatients(F=49.04, p=.0001). The each class was significantly different(p<0.05). The mean hours of direct nursing care of several direct nursing care activities in each class were increased according to increasing nursing care requirements of the inpatients(p<0.05) ; class III and class IV for medication and education ${\cdot}$ communication, class I, class III and class IV for measurement ${\cdot}$ observation, class I, class II and class IV for elimination ${\cdot}$ irrigation, all of class for exercise ${\cdot}$ safety. 3. Indirect nursing care activities and personnel time : Recognization 24.2%, house keeping activity 22.7%, charting 17.2%, personnel time 11.8% etc. The mean hours of indirect nursing care and personnel time per nursing manpower was 4.7 hrs. The mean hours of indirect nursing care and personnel time per duty were 294.8 min for day duty, 212.3 min for evening duty, 387.9 min for night duty, 143.3 min for shift duty. The mean of indirect nursing care hours and personnel time of night duty was longer than that of the other duty. 4. The mean hours of indirect nursing care and personnel time per patient per day was 2.5 hrs. 5. The mean hours of nursing care per patient per day in each class were class I 5.6 hrs, class II 6.4 hrs, class III 7.2 hrs, class IV 7.7 hrs. 6. The elements of the nursing care cost were composed of 2,212 won for direct nursing care cost, 267 won for direct material cost and 307 won for indirect cost. Sum of the elements of the nursing care cost was 2,786 won. 7. The mean cost of the nursing care per patient per day in each class were 15,601.6 won for class I, 17,830.4 won for class II, 20,259.2 won for class III, 21,452.2 won for class IV. As above, using modified hospice patient classification indicators and nursing care activity details, many critical ill patients were hospitalized in the hospice unit and it reflected that the more nursing care requirements of the patients, the more direct nursing care hours. Emotional ${\cdot}$ spiritual care, pain ${\cdot}$ symptom control, terminal care, education ${\cdot}$ communication, narcotics management and delivery, attending funeral ceremony, the major nursing care activities, were also the independent hospice service. But it is not compensated by the present medical insurance system. Exercise ${\cdot}$ safety, elimination ${\cdot}$ irrigation needed more nursing care hours as equal to that of intensive care units. The present nursing management fee in the medical insurance system compensated only a part of nursing car service in hospice unit, which rewarded lower cost that that of nursing care.

  • PDF

회사채 신용등급 예측을 위한 SVM 앙상블학습 (Ensemble Learning with Support Vector Machines for Bond Rating)

  • 김명종
    • 지능정보연구
    • /
    • 제18권2호
    • /
    • pp.29-45
    • /
    • 2012
  • 회사채 신용등급은 투자자의 입장에서는 수익률 결정의 중요한 요소이며 기업의 입장에서는 자본비용 및 기업 가치와 관련된 중요한 재무의사결정사항으로 정교한 신용등급 예측 모형의 개발은 재무 및 회계 분야에서 오랫동안 전통적인 연구 주제가 되어왔다. 그러나, 회사채 신용등급 예측 모형의 성과와 관련된 가장 중요한 문제는 등급별 데이터의 불균형 문제이다. 예측 문제에 있어서 데이터 불균형(Data imbalance) 은 사용되는 표본이 특정 범주에 편중되었을 때 나타난다. 데이터 불균형이 심화됨에 따라 범주 사이의 분류경계영역이 왜곡되므로 분류자의 학습성과가 저하되게 된다. 본 연구에서는 데이터 불균형 문제가 존재하는 다분류 문제를 효과적으로 해결하기 위한 다분류 기하평균 부스팅 기법 (Multiclass Geometric Mean-based Boosting MGM-Boost)을 제안하고자 한다. MGM-Boost 알고리즘은 부스팅 알고리즘에 기하평균 개념을 도입한 것으로 오분류된 표본에 대한 학습을 강화할 수 있으며 불균형 분포를 보이는 각 범주의 예측정확도를 동시에 고려한 학습이 가능하다는 장점이 있다. 회사채 신용등급 예측문제를 활용하여 MGM-Boost의 성과를 검증한 결과 SVM 및 AdaBoost 기법과 비교하여 통계적으로 유의적인 성과개선 효과를 보여주었으며 데이터 불균형 하에서도 벤치마킹 모형과 비교하여 견고한 학습성과를 나타냈다.