• Journal of Korean Society for Quality Management
• /
• v.47 no.3
• /
• pp.537-552
• /
• 2019

Purpose: The development of information technology makes it easy to utilize high-dimensional categorical data. In this regard, the purpose of this study is to propose a novel method to select the proper categorical variables in high-dimensional categorical data. Methods: The proposed feature selection method consists of three steps: (1) The first step defines the goodness-to-pick measure. In this paper, a categorical variable is relevant if it has relationships among other variables. According to the above definition of relevant variables, the goodness-to-pick measure calculates the normalized conditional entropy with other variables. (2) The second step finds the relevant feature subset from the original variables set. This step decides whether a variable is relevant or not. (3) The third step eliminates redundancy variables from the relevant feature subset. Results: Our experimental results showed that the proposed feature selection method generally yielded better classification performance than without feature selection in high-dimensional categorical data, especially as the number of irrelevant categorical variables increase. Besides, as the number of irrelevant categorical variables that have imbalanced categorical values is increasing, the difference in accuracy between the proposed method and the existing methods being compared increases. Conclusion: According to experimental results, we confirmed that the proposed method makes it possible to consistently produce high classification accuracy rates in high-dimensional categorical data. Therefore, the proposed method is promising to be used effectively in high-dimensional situation.

• The Korean Journal of Applied Statistics
• /
• v.31 no.6
• /
• pp.721-732
• /
• 2018

Latent class analysis (LCA) is an important tool to explore unseen latent groups in multivariate categorical data. In practice, it is important to select a suitable set of variables because the inclusion of too many variables in the model makes the model complicated and reduces the accuracy of the parameter estimates. Dean and Raftery (Annals of the Institute of Statistical Mathematics, 62, 11-35, 2010) proposed a headlong search algorithm based on Bayesian information criteria values to choose meaningful variables for LCA. In this paper, we propose a new variable selection procedure for LCA by utilizing posterior probabilities obtained from each fitted model. We propose a new statistic to measure the adequacy of LCA and develop a variable selection procedure. The effectiveness of the proposed method is also presented through some numerical studies.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.06a
• /
• pp.270-273
• /
• 2001

본 논문에서는 실시간 성능 관리를 위해 패킷 분석 시스템을 설계하고 구현하였다. 기존의 MIB 정보를 이용한 망 관리에서는 관리국의 주기적인 요청으로 각 에이전트의 MIB 정보를 가져와 분석하는 방식으로, 실시간 감시에는 적합하지 않은 단점이 있다. 따라서, 본 논문에서는 실시간 트래픽 감시를 위해 시스템을 설계하고 구현하였다. 제안된 시스템은 트래픽 상태를 감시하는 모니터링 시스템과 관측된 트래픽을 보여주는 인터페이스 부분으로 나눌 수 있다. 모니터링 시스템은 각 노드의 트래픽을 감시하여 각 패킷별로 구분하여 사용자 인터페이스에 넘겨주게 되며, 이를 사용자 인터페이스에서는 수치형 자료로 표시하거나, 범주형 자료인 그래프로 나타내게 된다. 이 시스템은 각 노드의 부하 여부를 감시하여, 비정상적인 트래픽의 폭주를 발견하게 되면 분석 모듈의 작동에 의해 해킹을 비롯한 네트워크 장애를 감지할 수 있다. 이는 실시간 망 관리의 중요한 기본 기술로 여러 분야에 활용할 수 있다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.05a
• /
• pp.217-220
• /
• 2003

본 논문에서는 특정 매개변수의 입력 없이 속성(attribute)에 따른 목적속성(class)값의 분포를 고려하여 연속형(conti-nuous) 값을 범주형(categorical)의 형태로 변환시키는 새로운 방법을 제안하였다. 각각의 속성에 대해 목적속성의 분포를 1차원 공간에 사상(mapping)하고, 각 목적속성의 밀도, 다른 목적속성과의 중복 정도 등의 기준에 따라 구간을 군집화 한다. 이렇게 생성된 군집들은 각각 목적속성을 예측할 수 있는 확률적 수치에 기반한 것으로, 각 속성이 제공하는 정보의 손실을 최소화하는 이산화 경계선을 갖고 있다. 제안된 데이터 이산화 방법의 향상된 성능은 C4.5 알고리즘과 UCI Machine Learning Data Repository 데이터를 사용하여 확인할 수 있다.

• The Korean Journal of Applied Statistics
• /
• v.18 no.1
• /
• pp.173-182
• /
• 2005

A Bayesian threshold model is considered to analyze binary or ordered categorical traits. Gibbs sampler for making full Bayesian inferences about the category probability as well as the regression coefficients is described. The model can be regarded as an alternative to the ordered logit regression model. Numerical examples are shown to demonstrate the efficiency of the model.

• Journal of Korean Society of Transportation
• /
• v.16 no.3
• /
• pp.93-100
• /
• 1998

지난 30여년간 운전자의 속도선택의 행태에 대하여 많은 연구가 이루어졌다. 그러 나, 과거 대부분의 연구는 운전자의 개별적인 특성과 제한속도에 대한 운전자의 인지 정도 를 고려하지 않고, 다만 운전자의 속도선택과 도로 및 차량간의 상호 관련성에 중점을 두고 있다. 본 연구는 운전자, 차량 및 통행특성 등의 요인을 고려하여 운전자의 속도 선택에 대 한 행태를 분석하고자 하였다. 이를 위하여 운전자의 속도 자료와 설문자료를 조사한 수, 두 가지 자료를 범주형 자료로 구분하여 Ordered Probit Model을 적용하여 분석하였다. 분 석결과 i) 고소득의 남성운전자가 고속의 주행 행태를 보였으며, 운전경력이 많은 운전자일 수록 높은 속도를 선택하는 것으로 나타났다. ii) 차량에 관해서는 배기량이 높은 차량일수 록 고속의 속도를 나타낸 반면에 안전장치가 많은 차량의 경우에는 저속의 주행속도를 보이 는 것으로 나타났다. iii) 통행 특성 면에서는 일일통행거리가 중요 변수인 것으로 나타났다. iv) 운전자의 심리적 측면에서는 운전자가 인식하고 있는 제한 속도가 또한 중요변수로 분 석되었다.

• Journal of the Korea Convergence Society
• /
• v.11 no.5
• /
• pp.153-158
• /
• 2020

The present study identify the predictive power of confrontational naming and generative naming as screening tests for normal and early cognitive impairment. The subjects were analyzed for 203 healthy elderly, 106 mild cognitive impairment (MCI), 31 mild dementia. The confrontational naming was measured by the short-term Korean Boston Name Waiting Test, and the generative naming was measured by the Control Associative Word Test. As a result of polynomial logistic regression, both confrontational naming and generative naming had a significant effect on discriminating cognitive impairment (MCI, mild dementia) in general elderly (p<0.05). On the other hand, when distinguishing mild dementia from mild cognitive impairment, the generative naming-phonetic test had no significant odds ratio. The results of this study suggest that when discriminating mild dementia in mild cognitive impairment group, it is not meaningful to look only at the total score of generative naming test.

• Journal of The Korean Association For Science Education
• /
• v.31 no.8
• /
• pp.1121-1144
• /
• 2011

This research is an investigation of difficulties encountered by authors of 5th and 6th grade science textbooks. The aim is to assist authors in creating more easily understandable textbooks in the future. In-depth interviews were conducted with 6 teachers who have previously taken part in the development of 5th and 6th grade texts. The responses given during these interviews were analyzed using open, axial, and selective coding as suggested by Strauss and Corbin (1998). The results are as follows: In open coding, related concepts were extracted and classified into 15 main categories and 46 sub-categories. In axial coding, the main categories were arranged into causal conditions, main phenomenon, context, intervening conditions, action and interactional strategies, and consequences wherein they were consistently related to each other based on Grounded theory. Finally, in selective coding, core categories were instilled whereby the texts being developed were categorized into conservative, progressive, and innovative to allow for easier interpretation. This was done to improve the overall quality of Science textbooks.

• Korean Journal of Social Welfare
• /
• v.68 no.4
• /
• pp.25-51
• /
• 2016

The purpose of this study was to explore the social enterprises work experience of marriage immigrant women. In-depth interviews with 10marriage immigrant women who were work at social enterprises in 'C' city were performed and analyzed by grounded theory method to configured 113 concepts and classified into 28 sub-categories 13 categories. The Central phenomenon was 'The resurgence of existence'and the Core category was 'Through the encounters of the institutional opportunities of social enterprise, experiencing a resurgence of existence and extending their presence as a Korean'. Also, Work experience of immigrant women working in social enterprises has been classified as a piggyback type, self-expandable type, co-prosperity type and situational model was presented. Based on the result, practical and policy proposal for marriage immigrant women's social enterprise employment were suggested.

• Korean Journal of Social Welfare
• /
• v.65 no.3
• /
• pp.155-182
• /
• 2013

The purpose of this study is to examine how the long-term care effects the elderly that adapt themselves to the caring facility, what their experiences are and what kinds of behavioral characteristics they present. We have analyzed the research conducted on 15 elderly individuals who are living in an elderly long-term care facility, by using the grounded theory approach of Strauss and Corbin (1998). As a result, 170 concepts, 42 sub-categories, and 15 categories were set by the open coding process. During the adaptation process in a long-term care facility, the primary experience or feeling by the elderly is that they had been 'deserted'. However, when consolidating the casual conditions, contextual conditions, intervening conditions, the action/interaction strategy, and consequence, the primary experience was that the elderly came to an 'acceptance'. Such acceptance was then sub-categorized into a destiny-resignation type, reality-acceptance type, and voluntary-selection type. Based on the results of this study, we recommend practical alternatives which will improve surrounding circumstances including caring facilities, its employees, relationships with other elderly individuals, and family support.