• Journal of Computing Science and Engineering
• /
• 제5권2호
• /
• pp.150-160
• /
• 2011

Automatic text classification has a long history and many studies have been conducted in this field. In particular, many machine learning algorithms and information retrieval techniques have been applied to text classification tasks. Even though much technical progress has been made in text classification, there is still room for improvement in text classification. In this paper, we will discuss remaining issues in improving text classification. In this paper, three improvement issues are presented including automatic training data generation, noisy data treatment and term weighting and indexing, and four actual studies and their empirical results for those issues are introduced. First, the semi-supervised learning technique is applied to text classification to efficiently create training data. For effective noisy data treatment, a noisy data reduction method and a robust text classifier from noisy data are developed as a solution. Finally, the term weighting and indexing technique is revised by reflecting the importance of sentences into term weight calculation using summarization techniques.

• Communications for Statistical Applications and Methods
• /
• 제30권1호
• /
• pp.1-19
• /
• 2023

Observational data with missing or incomplete data are common in biomedical research. Multiple imputation is an effective approach to handle missing data with the ability to decrease bias while increasing statistical power and efficiency. In recent years propensity score (PS) matching has been increasingly used in observational studies to estimate treatment effect as it can reduce confounding due to measured baseline covariates. In this paper, we describe in detail approaches to competing risk analysis in the setting of incomplete observational data when using PS matching. First, we used multiple imputation to impute several missing variables simultaneously, then conducted propensity-score matching to match statin-exposed patients with those unexposed. Afterwards, we assessed the effect of statin exposure on the risk of heart failure-related hospitalizations or emergency visits by estimating both relative and absolute effects. Collectively, we provided a general methodological framework to assess treatment effect in incomplete observational data. In addition, we presented a practical approach to produce overall cumulative incidence function (CIF) based on estimates from multiple imputed and PS-matched samples.

• 대한예방한의학회지
• /
• 제21권1호
• /
• pp.107-119
• /
• 2017

Purpose : Recently, cancer has become a chronic disease that requires supervision because of early diagnosis and the development of therapeutic technology. As a result, cancer patients are interested in improving the quality of their lives besides the treatment of cancer itself. Therefore, it is necessary to conduct a qualitative research to understand the vivid experiences of cancer patients and structure their treatment experience. Among qualitative researches, grounded theory is developed based on the data collected in the field. The grounded theory research method is easy to analyze for the process and structure of the treatment experience. Therefore, the purpose of this study is to provide basic data on the integrated medical experience of cancer patients Methods : Participants were conveniently selected, and the criteria for selection were for those who had more than 1 month of hospitalization so that they could dictate their situation and experience in a meaningful manner. Data was collected through in-depth interviews and continued until the data were saturated through theoretical sensitivity and continuous comparison methods. The collected data were analyzed through the process of open-coding, axial coding, and selection coding, which are the research methods of grounded theory. Results : Cancer patients differ in their path, purpose, and attitudes depending on their respective situations and the internal and external resources of individuals. There is also a difference in the perception of their situation among the cancer patients and their families. Cancer patients were shown to recognize and cope with problems in the integrated medical treatment process, and have been classified into 6 different types after the results. Cancer patients showed positive changes in terms of physical, emotional, and lifestyle after their integrated medical treatment. Cancer patients perceived the integrated medical treatment process as a horizontal relationship structure and with diversity. Conclusions : The experience of integrated medical treatment of cancer patients is a process of rehabilitation that heals the body and restores life within the interaction of support system, contextual situation, and internal resources of the individual. Despite this, there is a need not only for the efforts of integrated medical service providers but also institutional support in the future with regards to the current weaknesses and points for improvement. In addition, there is a need for an objective criterion to measure the outcome of integrated medicine for the standardization of integrated medical services.

• 응용통계연구
• /
• 제32권2호
• /
• pp.173-185
• /
• 2019

과학적 연구에서 핵심적인 연구 주제 또는 가설은 대부분 인과적 질문(causal question)을 포함한다. 예를 들어, 전염병 예방을 위한 치료법의 효과 연구, 특정 정책의 시행으로 인한 효용(utility)의 평가에 대한 연구, 특정 사용자를 대상으로 노출된 광고의 종류에 따른 광고의 효과성에 대한 연구는 모두 인과 관계(causal relationship)의 추론이 요구된다. 이러한 인과 관계를 다루는 통계적 인과 추론(statistical causal inference)의 주요 관심사 중 하나는 모집단에 일종의 개입(정책 혹은 처치)을 적용한 후 개입의 효과를 정확하게 추정하는 것이다. 인과 추론은 임상실험과 정책결정에서 주로 이용되었으나, 이른바 빅데이터 시대의 도래로 가용한 관측자료가 폭발적으로 증가하였고 이로 인하여 인과 추론에 대한 잠재적 응용가치와 수요가 지속적으로 증가하고 있다. 하지만 가용한 대부분의 자료는 임의실험 기반의 자료와 달리 개입이 임의로 분배되지 않은 비실험 관측자료이다. 따라서, 본 논문은 비실험 관측자료로부터 개입의 효과를 추정하기 위한 인과 추론의 핵심 개념과 최근의 연구동향을 소개하고자 한다. 이를 위하여 본문에서는 먼저 개입의 효과를 Neyman-Rubin의 잠재 결과(potential outcome) 모형으로 나타내고, 개입의 효과를 추정하는 여러 접근법 중 특히 성향점수(propensity score) 기반 추정법과 회귀모형 기반 추정법을 중점적으로 소개한다. 최근 연구동향으로는 (1) 평균 효과 크기 추정을 넘어선 개인별 효과 크기의 추정, (2) 효과크기 추정에 있어서 자료 규모의 증대로 인한 차원의 저주가 야기하는 난제들과 이에 대한 해결방안들, (3) 복합적 인과관계를 반영하기 위한 Pearl의 구조적 인과 모형(structural causal model) 및 잠재 결과 모형과의 비교의 3가지 주제로 구분하여 소개한다.

• 종양간호연구
• /
• 제1권1호
• /
• pp.32-43
• /
• 2001

High risk breast cancer patients receive aggressive treatments such as APBSCT in order to prevent the recurrence. These treatments take a long time and decrease the quality of life of patients as a result. The purpose of this study is to examine when the changes of quality of life are dramatic in general and in specific dimension. This is a time series study of two types of treatment groups, in which 15 patients were treated by chemotherapy only and nine patients by auto-peripheral blood stem cell transplantation. Subjects were in this study were stage III and above of breast cancer. Data were collected from April 1999 to April 2000 with the 3 month-interval starting after 3rd cycle of chemotherapy. Data were collected 3 times ; treatment period, treatment finishing period and 3 months later after the treatment. Ferrell's QOL instrument was used which had been developed for the breast cancer patients in 1989. Repeated Measure ANOVA was used to examine differences of quality of life at 3 points of time respectively. As a result, average age was 43.29 years (4.38yrs) and the number of married person was 22(91.7%). Scores of quality of life were 5.45 at 1st period after 3rd cycle of chemotherapy, 5.17 after treatment, 5.10 at 3 months later after treatment and difference of quality of life was decreased according to period of treatment (P=.085). Only the psychological dimension of the quality of life showed the significant difference (P=.045). Two different treatment groups showed a difference of quality of life at 3 month later after treatment. In conclusions, high risk breast cancer patients showed the decreased quality of life related to treatment up to 3 months later of treatment. Subjects who received APBSCT returned to the normal quality of life more rapidly than those who received chemotherapy. Psychological nursing intervention was needed during the whole period of treatment.

• Journal of Hospice and Palliative Care
• /
• 제26권3호
• /
• pp.112-125
• /
• 2023

Purpose: This study investigated knowledge, attitudes, and nursing stress related to life-sustaining treatment among oncology nurses. Methods: A descriptive study design was used. Data were collected through a survey from April 1 to May 31, 2022. The participants were 132 nurses working in the oncology ward of a tertiary hospital in Seoul. Data were analyzed using the SPSS 25.0 program with descriptive statics, the independent t-test, analysis of variance, and Pearson correlation coefficients. Results: The average scores for knowledge, attitudes, and nursing stress related to life-sustaining treatment were 14.42, 3.29, and 3.96, respectively. Significant differences in knowledge about life-sustaining treatment were observed based on clinical experience (P=0.029) and education about life-sustaining treatment (P=0.044). Attitudes toward life-sustaining treatment varied significantly with education about life-sustaining treatment (P=0.014), while stress levels differed significantly across working units (P=0.004). A positive correlation was found between the dilemma of extending or stopping life-sustaining treatment (a subdomain of nursing stress) and attitudes toward life-sustaining treatment (r=0.260, P=0.003). Conclusion: There was no significant correlation between the nursing stress experienced by oncology nurses and their knowledge and attitudes toward life-sustaining treatment. However, a more positive experience with life-sustaining treatment education was associated with higher stress levels related to the dilemma of extending or stopping life-sustaining treatment. Therefore, it is crucial to develop strategies to manage this dilemma and reduce stress in the field.

• 한국물환경학회지
• /
• 제30권5호
• /
• pp.524-531
• /
• 2014

Korea has been trying to manage water quality of rivers and lakes in many ways. Ministry of Environment is making continuous efforts of operation control improvement such as strengthening effluent standards in sewage treatment plants and expanding Tele-Monitoring System (TMS). However, evaluation method for operations of sewage treatment plants and establishment system of effluent standards are inadequate. The objective of this study is to evaluate for operations of sewage treatment plants using statistics technique such as frequency analysis, percentile, normal distribution analysis. We used the effluent data (BOD, COD, SS) collected at 299 plants in 2012. The values were very low by comparison with standards. The Data followed a normal distribution. We think that distribution characteristics are closely related with effluent standards, especially T-P and BOD. Statistics technique attempted in this study can be used to evaluate for operations of sewage treatment plants and assess the appropriateness of effluent standards based on TBEL(Technology-based effluent limitation). And, this technique can be used to figure out the overall level of plants, the status of each plant and the favorable treatment process for each item (BOD, COD etc.). Data acquired through this method can be used to improve facilities and operation techniques and decide on a sewerage policy.

• 한국융합학회논문지
• /
• 제9권2호
• /
• pp.101-109
• /
• 2018

본 연구는 하부요로증상을 가진 성인 남자의 치료 추구 행위 정도를 파악하고 치료추구행위의 영향 요인을 확인하고자 실시되었다. 연구는 공공기관에 근무하거나 공공기관을 방문한 성인 남자 중 하부요로증상을 가진 142명을 대상으로 2016년 10월에서 11월 사이에 구조적 설문지를 이용하여 이루어졌다. 수집된 자료는 SPSS/WIN 18.0 프로그램을 이용하여 분석하였으며 분석결과, 하부요로증상과 관련한 치료추구 행위를 한 대상자는 55.6%(79명), 치료추구행위를 하지 않았다고 응답한 대상자는 44.4%(63명)로 나타났다. 하부요로증상 관련 치료추구행위의 영향요인은 경제 수준, 주관적 건강상태, 자율성, 관계성 등이 확인되었다. 이는 하부요로증상을 가진 성인 남자의 건강증진을 위하여 치료추구행위를 촉진하는 전략마련이 시급함을 의미하며, 이러한 결과를 고려하여, 중재프로그램 을 개발한다면 대상자의 건강증진에 기여할 것으로 기대한다.

• 한국치위생학회지
• /
• 제17권5호
• /
• pp.875-887
• /
• 2017

Objectives: The objective of this study is to provide basic data needed in developing an educational program designed to upgrade capacity and awareness of preventive dental treatment among oral health workers, by analyzing levels of awareness of preventive dental treatment and educational needs among dentists and dental hygienists. Methods: The collected data was analyzed with SPSS program ver. 19.0. The data was under t-test. Results: The frequency level of giving preventive dental treatment to patients among dentists and dental hygienists is below mid-point, 3 on the 5-point Likert scale. In terms of frequency level per item, scaling & polishing was ranked the highest, followed by periodontal maintenance, tooth-brushing instruction, and prescription and instruction of oral care product in descending order. On the questions asking how important preventive dental care they perceive to be, both dentists and dental hygienists perceived it to be highly important. When they were asked to rank those items by the importance of education, they considered periodontal maintenance as the most important one, followed by individual education of oral health, incremental oral health care, scaling& polishing, toothbrushing instruction, and prescription and instruction of oral care product. Respondents pointed out problems in running a preventive dental treatment program as follows: overwork, lack of dedicated workforce, un-fixed costs, and lack of necessary equipment. When they were asked to point out items needed to run such a program, the largest number of respondents indicated dedicated workforce placement, followed by improving awareness of the customer, and improving awareness of the dental workers. Conclusions: In order to effectively run a preventive dental treatment program, it is necessary for oral health workers to clearly understand the concept of it. It is also necessary to develop and operate an education program on preventive dental treatment targeting oral health professionals.

• 한방재활의학과학회지
• /
• 제29권1호
• /
• pp.31-39
• /
• 2019

Objectives To obtain future research basis of Korean Medicine for total knee arthroplasty patient by analyzing medical and Korean Medical service utilization and treatment duration. Methods Data sampling was performed on Health Insurance Review and Assessment Service patient data of 2015 (Confidence level of 97%) to analyze patients' medical and Korean Medical service tendency. Sampling groups were divided into two groups; i) Patients who completed their treatment within 5 months of total knee arthroplasty, ii) Patients who continued their treatment after 5 months of total knee arthroplasty, to investigate patients' medical and Korean Medical service tendency and individual characteristics were carefully monitored. Results It was confirmed that total of 1,655 patients had gone through total knee arthroplasty out of 1,453,486 patients who were gathered for sampling. First sampling group (Patients who completed their treatment within 5 months of total knee arthroplasty) was 287 patients and second sampling group (Patients who continued their treatment after 5 months of total knee arthroplasty) was 385 patients. The proportion of patients who visited Korean Medical service in first sampling group was lower than that of second sampling group. Conclusions It was confirmed that medical and Korean Medical service and cost consumed by second group (Patients who continued their treatment after 5 months of total knee arthroplasty) was higher than that of first group (Patients who continued their treatment after 5 months of total knee arthroplasty). It is highly recommended to continue with further study for efficient medical and Korean Medical service and reduced cost.