• 자원환경지질
• /
• 제27권3호
• /
• pp.237-246
• /
• 1994

Tin mineralizations in South Korea have been found only in the Ulchin and Sangdong areas. They appear to be in close spatial association with the Wangpiri granitoid in the UlChin area, and the Nonggeori and Naedeogri granites in the Sangdong area. However, previous works have revealed that there are considerable differences in geological setting, mineralogical and geochemical compositions among these granitoids concerned. The roles of discriminant and multiple regression analysis have been examed to establish geochemical differences among the tin-granitoids and to identify elements relating to tin mineralizations. The data set used in this study consists of 60 observations with 29 elements which are cited from pre-existing publications. A stepwise discriminant analysis determined the group of variables that differentiate between samples from four training sets; Buncheon, Wangpiri, Nonggeori and Naedeogri granitoids. These granitoids were most effectively discriminated on the basis of major elements FeO, CaO and $P_2O_5$ and also by the trace elements Rb and Zr. Results of the multiple regression analysis shows that the level of Sn in granitoids depends positively on ones of MnO, Rb and FeO and negatively $P_2O_5$. Graphical representation of discriminant scores on sampling locations greatly aid recognition of differences in the geochemical characteristics in terms of spatial distribution of granitoids examed. The application of the discriminant analysis provides a potential means of identifying and comparing geochemical characteristics.

• 대한간호학회지
• /
• 제23권1호
• /
• pp.130-141
• /
• 1993

This study was to find out variables influenced on the medication patterns (voluntary medication, in-voluntary medication) of antipsychotics taken by schizophrenic outpatients. Purposes of this study were to be identified that there was the significant difference between the group of voluntary medication and involuntary, and that which variables had infuence on outpsychotics medication. The sample consisted of 30 patients takeing their pills voluntary (voluntary medication group), and 15 patients involuntary(involuntary medication group) at a psychiatry hospital and a psychiatric unit of a The findings of the study are as follows : university hospital in Daegu. Data were collected from September to October, 1991 through interview using questionare about antipsychotics medication. Data were analyzed by the technique of two group discriminant function analysis using SPSS pc$^{+}$ 1) Discriminant function discriminating between voluntary medication and involuntary medication was significant at the level of 10% significance (sig.=.055, p〈.10) Hit-ratio was very high (91. 1％) 2) One of General variables, SEX, was significant of discriminating between two medication patterns at the level of 10% significance. 3) One of Family Environmental Variables, BROTH(a number of brother), were significant between two medication patterns. (p〈.10) 4) One of Therapeutic Environmental Variables, FAMHX, was significant between two medication patterns at the level of 10% significance. 5) One of Variables Related to Drug and Medication, NECESS, was significant between two medication patterns. (p〈.05) 6) Variables Related to Disease was not significant between two medication patterns.s.

• 응용통계연구
• /
• 제21권3호
• /
• pp.455-468
• /
• 2008

야구경기의 승패에 영향을 미치는 중요한 요인으로 간주되는 도루의 성공모형을 개발하기 위하여 2007년 한국프로야구 기록자료를 바탕으로 로지스틱 회귀모형들을 제안한다. 또한 한국프로야구의 도루성공과 실패에 대해 판별분석을 실시하고 분류 기준값을 결정하였으며, 판별분석 분류표를 이용해 로지스틱 회귀분석과 판별분석의 효율성을 비교한다. 전체적인 모형의 정확도는 로지스틱 회귀모형이 판별분석보다 더 좋은 것으로 나타났고, 연속형 자료를 범주형으로 변환한 자료에 대한 로지스틱 회귀모형도 유사한 효율성을 갖고있다.

• 대한간호학회지
• /
• 제41권5호
• /
• pp.652-662
• /
• 2011

Purpose: The suicide rate of adolescents in Korea is increasing annually. Therefore, this research was done to identify the suicide attempt rate of middle and high school students and to identify factors that influence suicidal attempts. Methods: The Korea Youth Health Risk Behavior Web-based Survey (2007) was used as data. Discriminant analysis and logistic regression were performed to analyze the data depending on gender to consider the gender difference in assessing the influence of each independent variable on suicidal attempts. Results: Discriminant analysis according on gender showed that 13 factors correlated with suicidal attempts for boys, and 20 factors for girls. The most highly correlated factors were smoking, depression and inhalation experience. For inhalation experience, boys had 2.7 times higher possibility of suicide attempts (95% CI 1.8-3.0) and girls, a 2.4 times higher possibility (95% CI 1.7-3.5). Conclusion: The results of the study indicate a need to classify adolescents for expectation of suicide risk and high danger for suicidal attempts through, and introduce suicide prevention programs for these adolescents. In particular, it is necessary to start intervention with students who smoke, have sexual and inhalation experiences and high levels of depression.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권4호
• /
• pp.332-339
• /
• 2014

Face classification has wide applications in security and surveillance. However, this technique presents various challenges caused by pose, illumination, and expression changes. Face recognition with long-distance images involves additional challenges, owing to focusing problems and motion blurring. Multiple frames under varying spatial or temporal settings can acquire additional information, which can be used to achieve improved classification performance. This study investigates the effectiveness of multi-frame decision-level fusion with photon-counting linear discriminant analysis. Multiple frames generate multiple scores for each class. The fusion process comprises three stages: score normalization, score validation, and score combination. Candidate scores are selected during the score validation process, after the scores are normalized. The score validation process removes bad scores that can degrade the final output. The selected candidate scores are combined using one of the following fusion rules: maximum, averaging, and majority voting. Degraded facial images are employed to demonstrate the robustness of multi-frame decision-level fusion in harsh environments. Out-of-focus and motion blurring point-spread functions are applied to the test images, to simulate long-distance acquisition. Experimental results with three facial data sets indicate the efficiency of the proposed decision-level fusion scheme.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2003년도 추계학술대회
• /
• pp.57-72
• /
• 2003

This article provides integrated neural network models for the interest rate forecasting using change-point detection. The model is composed of three phases. The first phase is to detect successive structural changes in interest rate dataset. The second phase is to forecast change-point group with data mining classifiers. The final phase is to forecast the interest rate with BPN. Based on this structure, we propose three integrated neural network models in terms of data mining classifier: (1) multivariate discriminant analysis (MDA)-supported neural network model, (2) case based reasoning (CBR)-supported neural network model and (3) backpropagation neural networks (BPN)-supported neural network model. Subsequently, we compare these models with a neural network model alone and, in addition, determine which of three classifiers (MDA, CBR and BPN) can perform better. For interest rate forecasting, this study then examines the predictability of integrated neural network models to represent the structural change.

• Journal of the Korean Data and Information Science Society
• /
• 제17권2호
• /
• pp.269-278
• /
• 2006

This paper presents the new hybrid data mining technique using error pattern, modeling of improving classification accuracy. The proposed method improves classification accuracy by combining two different supervised learning methods. The main algorithm generates error pattern modeling between the two supervised learning methods(ex: Neural Networks, Decision Tree, Logistic Regression and so on.) The Proposed modeling method has been applied to the simulation of 10,000 data sets generated by Normal and exponential random distribution. The simulation results show that the performance of proposed method is superior to the existing methods like Logistic regression and Discriminant analysis.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1395-1405
• /
• 2017

In this paper, we analyses heart beat data to identify subjects stress state (binary) using heart rate variability (HRV) features extracted from heart beat data of the subjects and implement supervised machine learning techniques to create the mental stress classifier. There are four steps need to be done: data acquisition, data processing (HRV analysis), features selection, and machine learning, before doing performance measurement. There are 56 features generated from the HRV Analysis module with several of them are selected (using own algorithm) after computing the Pearson Correlation Matrix (p-values). The results of the list of selected features compared with all features data are compared by its model error after training using several machine learning techniques: support vector machine, decision tree, and discriminant analysis. SVM model and decision tree model with using selected features shows close results compared to using all recording by only 1% difference. Meanwhile, the discriminant analysis differs about 5%. All the machine learning method used in this works have 90% maximum average accuracy.

• Korean Chemical Engineering Research
• /
• 제46권2호
• /
• pp.383-388
• /
• 2008

화학공정의 안전하고 효율적인 운전에 관심이 커지면서 공정이상의 원인을 조기에 진단하기 위한 다양한 이상진단방법이 연구되어 왔다. 최근에는 통계적 모델 등 정량적 데이터에 기반한 이상진단방법이 많이 연구되고 있으나, 특정 조업영역에서 얻어진 통계적 모델을 다른 조업영역에 적용하면 오진단이 많아지게 된다. 따라서 공정특성상 다양한 조업영역이 존재하는 화학공정에 데이터기반 방법론을 적용하기에는 어려움이 있어 화학공정의 조업영역 판별법이 요구되고 있다. 이 연구에서는 유클리드 거리(Euclidean distance), FDA(Fisher's discriminant analysis), PCA(principal component analysis)의 통계모델과 이 모델들에 공정변수의 동특성을 반영한 모델을 제안하였다. 6개의 조업모드를 가진 TE(tennessee eastman) 공정에 대한 사례연구를 통해 동특성을 반영한 PCA 모델의 성능이 가장 우수함을 확인하였다.

• 한국지능시스템학회논문지
• /
• 제22권6호
• /
• pp.735-740
• /
• 2012

본 연구에서는 주성분 분석법 및 선형 판별 분석법을 이용한 다항식 방사형 기저 함수 신경회로망 분류기의 설계 방법론을 소개한다. 주성분 분석법과 선형판별 분석법을 사용하여 주어진 데이터의 정보 손실을 최소화한 특징데이터를 생성하고 이를 다항식 방사형 기저함수 신경회로망의 입력데이터로 사용한다. 방사형 기저 함수 신경회로망의 은닉층은 FCM 클러스터링 알고리즘으로 구성되며 연결가중치는 1차 선형식을 사용하였다. 최적의 분류기 설계를 위해서 최근에 제안된 Artificial Bee Colony(ABC) 최적화 알고리즘을 사용하여 구조 및 파라미터를 동조하였다. ABC 알고리즘을 통해 주성분 분석법과 선형판별 분석법의 고유벡터의 수 및 FCM 클러스터링 알고리즘의 퍼지화 계수등의 파라미터를 동조한다. 제안된 분류기는 대표적인 Machine Learning(ML) 데이터를 사용하여 성능을 평가하며 기존 분류기와 성능을 비교한다.