• Journal of the Korean Data and Information Science Society
• /
• 제26권2호
• /
• pp.517-524
• /
• 2015

Unlabeled examples are easier and less expensive to be obtained than labeled examples. In this paper semisupervised approach is used to utilize such examples in an effort to enhance the predictive performance of nonlinear quantile regression problems. We propose a semisupervised quantile regression method named semisupervised support vector quantile regression, which is based on support vector machine. A generalized approximate cross validation method is used to choose the hyper-parameters that affect the performance of estimator. The experimental results confirm the successful performance of the proposed S2SVQR.

• Journal of the Korean Statistical Society
• /
• 제33권1호
• /
• pp.25-34
• /
• 2004

In this paper we propose a prediction method on the regression model with randomly censored observations of the training data set. The least squares support vector machine regression is applied for the regression function prediction by incorporating the weights assessed upon each observation in the optimization problem. Numerical examples are given to show the performance of the proposed prediction method.

• 사물인터넷융복합논문지
• /
• 제8권5호
• /
• pp.127-132
• /
• 2022

사물인터넷의 발달에 따라 IoT 기기에서 처리되는 데이터의 정확성도 중요시되고 있다. 사물인터넷에서 생산되는 데이터는 센서마다 다양한 포맷과 프로토콜을 사용하고 있기에 수집된 센서 데이터에 이상이 발생하면 정규화하고 통합하는 과정에서 데이터 오류로 인해 실패하거나 잘못된 데이터를 구성하게 된다. 사용자의 상황이나 IoT 기기의 이상 증상은 정확하게 판단되지 않기 때문에 사용자 서비스 장애가 발생하거나 실패하는 문제가 발생한다. 본 논문은 IoT 환경에서 발생되는 다양한 형태의 데이터가 IoT 기기의 특성을 반영하여 정상적인 범주 내에서 변화되는지를 수학적 함수로 산출하여 데이터의 정합성을 탐지하는 방법을 제안한다. IoT 데이터의 발생 특성을 파악하기 위해 '기울기 분석'을 활용한 방법과 '선형 회귀 분석'을 활용한 방법을 각각 제시하고 실험하였다. 기울기를 활용하는 방법은 '증가하는 속도'가 다음에 일어나는 현상에 영향을 미치는 IoT 데이터(센서 기기)에 적합하며, 선형 회귀를 활용하는 방법은 선형적으로 데이터가 움직일 때 '선형 회귀 함수로부터의 차이'가 다음에 일어나는 현상에 영향을 미치는 데이터(수도, 전기 계량기)에 적합하였다.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2008년도 제30회 춘계학술대회논문집
• /
• pp.89-92
• /
• 2008

일반적으로 하이브리드 추진에서 산화제 질량유속만의 함수로 표현되는 후퇴율 식은 고체연료 길이에 따른 후퇴율 변화를 나타내지 못한다. 따라서 본 연구에서는 분할 연료 그레인을 적용해 쉽게 할 수 있는 연소 실험을 수행했고, 고체연료 길이에 따른 후퇴율의 변화를 고찰하였다. 연료 그레인 상단부에서 하단부로 갈수록 후퇴율은 감소하다가, 다시 증가하는 경향을 나타남을 확인 하였고, 산화제 질량유속과 그레인 길이의 함수로 표현할 수 있는 후퇴율 식을 도출하였다.

• 응용통계연구
• /
• 제26권1호
• /
• pp.141-149
• /
• 2013

로지스틱회귀모형에서 두 설명변수의 조건부 분포가 모두 이변량 정규분포라고 할 수 있다면 설명변수들의 함수로 표현되는 로그-밀도비를 통해 모형에 포함시켜야하는 항을 알 수 있다. 두개의 이변량 정규분포에서 분산-공분산행렬이 같은 경우에는 이차항과 교차항 없이 일차항만으로 충분하다. 상관계수가 모두 0이면 교차항은 설명변수의 분산과 관계없이 필요하지 않다. 또한 로지스틱회귀모형에서 로그-밀도비를 통해 이차항과 교차항이 필요하지 않게 되는 다른 조건들도 알아본다.

• Geomechanics and Engineering
• /
• 제10권3호
• /
• pp.269-284
• /
• 2016

Simplified techniques based on in situ testing methods are commonly used to assess seismic liquefaction potential. Many of these simplified methods were developed by analyzing liquefaction case histories from which the liquefaction boundary (limit state) separating two categories (the occurrence or non-occurrence of liquefaction) is determined. As the liquefaction classification problem is highly nonlinear in nature, it is difficult to develop a comprehensive model using conventional modeling techniques that take into consideration all the independent variables, such as the seismic and soil properties. In this study, a modification of the Multivariate Adaptive Regression Splines (MARS) approach based on Logistic Regression (LR) LR_MARS is used to evaluate seismic liquefaction potential based on actual field records. Three different LR_MARS models were used to analyze three different field liquefaction databases and the results are compared with the neural network approaches. The developed spline functions and the limit state functions obtained reveal that the LR_MARS models can capture and describe the intrinsic, complex relationship between seismic parameters, soil parameters, and the liquefaction potential without having to make any assumptions about the underlying relationship between the various variables. Considering its computational efficiency, simplicity of interpretation, predictive accuracy, its data-driven and adaptive nature and its ability to map the interaction between variables, the use of LR_MARS model in assessing seismic liquefaction potential is promising.

• 대한산업공학회지
• /
• 제30권1호
• /
• pp.27-35
• /
• 2004

A study was conducted to investigate an optimal vibration frequency for mobile phones with just noticeable difference(JND). The just noticeable difference, Weber's law, is the minimum amount by which stimulus intensity must be changed in order to produce a noticeable variation in sensory experience. In order to find the optimal vibration frequency, sixteen frequencies ranged from 24Hz to 603Hz were selected. Subjects then wereasked to differentiate a pair of vibration frequencies. For the analysis, the psychometric function to determine the optimal vibration frequency and the logistic regression to validate the determined frequency were used. The results show that the 2nd order polynomial equations were best fitted for the JND psychometric function and the optimal mobile phone vibrations were determined at 140Hz, 151 Hz, and 160Hz. With the ogive-shaped psychometric function developed by the logistic regression, the results of this study was validated that the determined vibration frequencies (140Hz, 151 Hz, and 160Hz) were optimal mobile phone vibration frequencies.

• KIEAE Journal
• /
• 제9권1호
• /
• pp.107-113
• /
• 2009

The energy cost is resulted from the energy use. Its sources are divided into some types and depended on the building use or energy-use type. The energy cost should be affected by the amount of the energy use. The cost could be calculated to consider various factors such as the insulation, heating type, building shape and others. But it can not consider all of the affect factors to the energy cost and need to categorize the factors to the condition for estimating the cost. In this paper, it aimed at providing the estimation model in linear equation and multiple linear regression, utilizing the building exterior condition and management characteristics in apartment housing. Its survey are conducted in two parts of management characteristics and building exterior condition. The correlation analysis is conducted to get rid of the multicolinearity among the inputted factors. The number of linear equation model is 11 and includes the 1st, 2nd and 3rd equation function, power function and others. Among these, it suggested the 2nd and 3rd function and power function in terms of the statistics. In multiple linear regression model, the building volume and management area are inputted to the estimation.

• 한국치위생학회지
• /
• 제19권1호
• /
• pp.65-78
• /
• 2019

Objectives: This study was conducted to characterize the impact of cognitive function and oral health status on mastication in senior citizens, ${\geq}65$ years of age, using senior centers in the city of Wonju, South Korea. Methods: A cross-sectional study consisting of a simple oral examination and survey questionnaires was performed in 154 individuals. General characteristics, subjective masticatory function, objective masticatory function, cognitive function, and oral health status were collected as variables. Correlation and multiple linear regression analyses were conducted. A p-value of <0.05 was considered to be statistically significant. Results: The subjective masticatory function was scored using the 5-point Likert scale. When subjective masticatory function was analyzed in groups according to cognitive function, the mean subjective masticatory function scores were 4.31, 4.09, and 3.29 in the normal group (cognitive score of ${\geq}16$), suspected dementia group (cognitive score of 1215), and mild dementia group (cognitive score of ${\leq}11$), respectively. Thus, subjective masticatory function decreased along with decreasing cognitive function. When cognitive function, subjective masticatory function, and objective masticatory function were compared with indicators of oral health status (number of functional teeth, oral dryness), subjective masticatory function exhibited a significant positive correlation with objective masticatory function (r=0.635, p<0.01), cognitive function (r=0.292, p<0.01), and total number of functional teeth, including prosthetic appliances (dentures) (r=0.305, p<0.01). According to the regression analysis, age, sex, number of functional teeth, and cognitive function affected subjective masticatory function. Conclusions: The results of this study revealed that age, sex, number of functional teeth, and cognitive function affected subjective masticatory function, whereas oral dryness did not. Therefore, dental professionals must consider subjective masticatory function when providing oral care in senior patients with low cognitive function.

• Child Health Nursing Research
• /
• 제14권1호
• /
• pp.61-70
• /
• 2008

Purpose: The purpose of this study was to investigate the relationship between high school students' self-differentiation, family function and their level of stress. Method: A questionnaire which consisted of questions on general characteristics of the high school students, and 36 questions on self-differentiation, 17 questions on family function, and 37 questions on level of stress was used to collect the data. Participants were 201 second grade high school students from Bucheon City. Descriptive statistics, T-test, ANOVA, correlation and multiple regression were used with SPSS 10.0 to analyze the data. Results: The mean scores for self-differentiation, family function, and levels of stress were 3.27, 3.39, and 2.61 respectively. The relationship between self-differentiation and level of stress revealed a significant negative correlation. The relationship between self-differentiation and family function showed a significant positive correlation. The relationship between family function and stress level showed a significant negative correlation. Conclusion: The results of the study show that variation in level of stress was related to family regression, recognition/emotional function, family projection, role recognition and emotional support and emotional cutoff which together explained 40.9% of the variance in level of stress.