RISS 학술연구정보서비스

검색
다국어 입력

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

변환된 중국어를 복사하여 사용하시면 됩니다.

예시)
  • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
  • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
닫기
    인기검색어 순위 펼치기

    RISS 인기검색어

      검색결과 좁혀 보기

      선택해제
      • 좁혀본 항목 보기순서

        • 원문유무
        • 원문제공처
        • 등재정보
        • 학술지명
        • 주제분류
        • 발행연도
        • 작성언어
        • 저자
          펼치기

      오늘 본 자료

      • 오늘 본 자료가 없습니다.
      더보기
      • 무료
      • 기관 내 무료
      • 유료
      • KCI우수등재

        단어 특징의 의미적 보강을 이용한 트위터 뉴스 분류 기법

        지선미(Seonmi Ji),문지훈(Jihoon Moon),김현우(Hyeonwoo Kim),황인준(Eenjun Hwang) 한국정보과학회 2018 정보과학회논문지 Vol.45 No.10

        최근 많은 사람들이 트위터를 뉴스 플랫폼으로 활용하면서 많은 뉴스 기사가 끊임없이 생성되고, 기사와 관련된 다양한 정보와 의견들이 빠르게 확산되고 있다. 그러나 트위터 뉴스는 동시다발적으로 포스팅되기 때문에 사용자가 원하는 주제의 기사를 선별하여 보기가 어렵다는 문제가 있다. 이를 위해, 트위터 뉴스를 주제별로 분류하기 위한 기계 학습과 딥러닝 기반의 다양한 연구들이 진행되었다. 하지만 통상적인 기계 학습 기법은 트위터 뉴스를 임베딩하는 과정에서 데이터 희소성이나 시맨틱 갭의 문제가 발생할 수 있으며, 딥러닝 기법은 많은 양의 데이터를 필요로 한다. 이러한 단점을 개선하기 위해, 본 논문에서는 적은 양의 데이터로도 데이터 희소성과 시맨틱 갭 문제를 해결할 수 있는 방법으로, 단어 특징의 의미적 보강을 이용한 트위터 뉴스 분류 기법을 제안한다. 구체적으로 먼저, 벡터 공간 모델을 이용하여 수집된 트위터 뉴스 데이터의 특징을 추출하고, DBpedia Spotlight를 통해 DBpedia의 자원과 온톨로지 정보를 반영하여 추출된 특징을 보강한다. 보강된 특징 단어 집합을 이용하여 다양한 기계 학습 기법 기반의 주제별 분류모델을 구성하고, 다양한 실험을 통해 제안하는 모델이 기존의 기법들보다 더 효과적임을 보인다. Recently, with the popularity of Twitter as a news platform, many news articles are generated, and various kinds of information and opinions about them spread out very fast. But since an enormous amount of Twitter news is posted simultaneously, users have difficulty in selectively browsing for news related to their interests. So far, many works have been conducted on how to classify Twitter news using machine learning and deep learning. In general, conventional machine learning schemes show data sparsity and semantic gap problems, and deep learning schemes require a large amount of data. To solve these problems, in this paper, we propose a Twitter news-classification scheme using semantic enrichment of word features. Specifically, we first extract the features of Twitter news data using the Vector Space Model. Second, we enhance those features using DBpedia Spotlight. Finally, we construct a topic-classification model based on various machine learning techniques and demonstrate by experiments that our proposed model is more effective than other traditional methods.

      • KCI등재후보

        일차의료 만성질환관리 시범사업 소개 및 효과

        송은주 ( Eunju Song ),김영은 ( Young-eun Kim ),지선미 ( Seonmi Ji ) 대한내과학회 2021 대한내과학회지 Vol.96 No.1

        Considerable efforts have been made to prevent and manage hypertension and diabetes, but the prevalence of these diseases has not changed significantly over the past 10 years. Hypertension and diabetes are chronic conditions requiring continuous treatment and management. Controlling blood pressure and blood sugar levels can prevent life-threatening complications and a deterioration in the quality of life. Community-centered primary health care is characterized by inclusiveness and continuous contact. Well-established primary health care services can reduce the number of unnecessary emergency room visits, hospitalizations, and medical tests, and therefore minimize unnecessary medical expenses. However, the number of patients treated for mild chronic diseases at large hospitals has increased, with a concomitant decrease in the proportion of outpatient clinic visits (41.6% in 2007 vs. 37.8% in 2016). Thus, there is a clear need to strengthen primary health care, including for the treatment of chronic diseases. To establish a chronic diseases management system based on local clinics, the Ministry of Health and Welfare of Korea launched a pilot chronic diseases management program aimed at increasing patient acceptance of local primary health care. The program was implemented in 2014, 2016, and 2019. This review examines the short-term impact of the pilot program with respect to the prevention and management of hypertension and diabetes. (Korean J Med 2021;96:7-12)

      • SCOPUSKCI등재

        우리나라 자살률의 추세변동에 관한 연구

        박종순(Jongsoon Park),김순덕(Soonduck Kim),지선미(Seonmi Ji),이제숙(Jesuk Lee) 한국역학회 2003 Epidemiology and Health Vol.25 No.2

        Purpose :This study was conducted to analysis the trends and states of suicide mortality in Korea from 1983 to 2001. Methods: This study was based on data published in 1983-2001 from the annual report of the cause of death statistics of National Statistical Office. Results: First, both of annual suicide rate and adjusted suicide rate rapidly went up in 1996 and hit the highest in 1998. Second, since 1983 only when the suicide rate was the highest in the ages 20-24 rather than other age groups, the suicides have been continuously shown to be the highest in the old age group. Third, the seasonal percentages of suicides were generally high in spring and summer, the seasons when the percentage of sunshine is relatively high, in both results for men and women. Fourth, men have most frequently committed suicide through hanging while their suicide through falling from a high site have least frequently occurred but currently tends to gradually increase. Meanwhile, women have most frequently done suicide generally by using insecticide. Conclusion: In Korea, the suicide rate and suicide percentage of the old people have been yearly increasing, and their suicide rate is also shown to the highest rather than those of other age groups. Therefore, it is necessary to prevent the suicide of the old people through a systematic analysis. Moreover, it is necessary to verify the risk factors and defensive factors for the suicide impulse and suicide attempt. On the basis of this, the construction of a nationwide social safety network for the national mental health is very important to find and treat the problems of the mental health to lead the people to suicide.

      • SCOPUSKCI등재

        서울시 일부 보건소에 등록된 폐결핵 재발 환자의 특성에 관한 연구

        한송이(Songyi Han),김순덕(Soonduck Kim),박종순(Jongsoon Park),지선미(Seonmi Ji),이제숙(Jesuk Lee) 한국역학회 2003 Epidemiology and Health Vol.25 No.2

        Purpose: The purpose of this study is to survey general and clinical characteristics of a group of relapsed patients with pulmonary tuberculosis, who are registered at a public health center in order to provide basic data for treatment of a patient s relapse, follow-up examination after the flrst treatment and health education. Subject: Among 68 relapsed patients with pulmonary tuberculosis registered at two public health centers in Seoul from Jan, I, 2002 to Dec. 31, 2002, 50 patients were selected as research subjects because they were contactable by telephone and they agreed to be a research subject. Method: The research carried out the structured questionnaire survey by telephone and surveyed research subjects' general and clinical characteristics, diseases except tuberculosis, drinking and smoking habit and knowledge about pulmonary tuberculosis treatment. Furthermore, the results of x-ray, whether a cavity existed or not, bacteriologic test, culture examination and drug resistance examination were examined and interval to relapse and history of dose attitude were surveyed. Results: Among relapsed tuberculosis patients, male patients were 60% and they were more than women(40%). Patients over 60 were 42% most. ]n their occupation, 30% of them were unemployed and 50% of patients surveyed had monthly income below one million won. According to the findings in radiographic extent, most 50% of them were minimal, 44% were moderately advanced and 2% were far advanced. Patients who had cavity were 30%. According to the result in bacteriologic test, 54% showed positive smear pulmonary tuberculosis and 50.0% positive culture. According to the result in drug resistance, 24% showed drug resistance against one drug and 76% showed sensitive response. 46% had experienced medication side effects. According to interval to a relapse, 32% were relapsed less than 2 years, 14% less than 3-5 years and 54% more than 5 years. Average interval before a relapse were 11.6 years(11.6±11.15). There was a difference between their past regimen and their present regimen(P=0.001). The attitude towards taking a medication was assessed and they showed higher compliance with medication at present than in the past(P=0.001). In a case of smoking rate during treatment, the rate of non-smokers, who had been smokers, were higher(P=0.047). Conclusion: As a result, in order to prevent the disease from being relapsed and to increase the rate of treatment effectiveness, we need to make patients recognize the importance of regular follow-up examination and we should control thoroughly the patients. Furthermore, we need to provide them with more health education not to stop taking a medication earlier.

      연관 검색어 추천

      이 검색어로 많이 본 자료

      활용도 높은 자료

      해외이동버튼