• 제목/요약/키워드: upper-bound method

검색결과 342건 처리시간 0.026초

홧병환자의 한의학적 치료에 대한 임상적 연구 (A Clinical Study on Treatments of Hwabyung with Oriental Medicine)

  • 김종우;황의완
    • 대한한의학회지
    • /
    • 제19권2호
    • /
    • pp.5-16
    • /
    • 1998
  • Hwabyung is a common emotional disorder which has symptoms expressed like firt's explosion in middle-aged after long period of emotional suppression among Koreans. It is similar in its characteristics such as neurosis, anxiety, panic attacks in Western Medicine, though the treatment method was not effective. So we have done a clinical research on Oriental Medical Method, especially on Acupuncture Therapy, and obtained following results. 1. Patients with Hwabyung complained of pressure pain around the Chunjung(?中, CV-17) point distinctively. About 70% of those were located on the CV-17 point, 25% were 1cm upper than the CV-17 point and 5% of those were 1cm lower point than the CV-17 point. 2. Degrees of pressure pain were divided into 5 grades from ade 1(feeling pain with slight pressure) to grade 5(feeling no pain with severe pressure), respectively. 3. Patients with Hwabyung showed various symptoms compared to fire's explosion such as anger, chest discomfort, difficulty in breathing. tachycardia. and feeling of epigasfric mass etc., and the degrees were divided into 5 grades according to the severities from grade 1(can't keep their usual living) to grade 5(no complaints with heavy stresses), respectively. 4. For the treatment of Hwabyung in this study, we had given Acupuncture therapy on some points such as Chunjung:?中:CV-17, Jungwan:中脘:CV-12) and Chunchu:天樞:S-25, etc. for 15 minutes a time twice a week. And Bunshimkiumgmnihang(分心氣飮加味方) was administered 3 times a day. 5. About 40% of the patients took treatment for more than 2 months, 29% of those took 1 to 2 months and 31% of those took less than 1 month. In this study, we excluded those who stopped treatment within a month without any expected effects. 6. We evaluated the changes of severity of pain according to the following categories such as - for no change, + for 1 grade, ++ for 2 grades, +++ for 3 grades, and ++++ for 4 grades of improvements. Among the patients taken 1 to 2 months of treatment. 48% of the those showed +, 7% of those showed ++, 3% of those showed +++ and 41% of those showed no change. Among the patients taken less than 2 months of treatment, 20%of those showed +, 40% of those showed ++, 28% of those showed +++ and 13% of those showed no change. 7. We evaluate the changes of symptoms according to the following categories such as - for no change, + for 1 grade, ++ for 2 grades, +++ for 3 grades and +++ for 4 grades of improvements. Among the patients taken 1 to 2 months of treatment, 34% of those showed +, 14% of those showed ++ and 52% of those showed no change. Among the patients taken more than 2 months of treatment, 20% of those showed +, 43% of those showed 20% of those showed +++, 3% of those showed +++ and 15% of those showed no change. 8. When we compare the changes of pain and symptoms according to the periods of treatment, the changes in quantity of pain in 1 to 2 months group was $0.72{\pm}0.75$, in more than 2 months group was $1.83{\pm}0.98$, and the changes in quantity of symptoms in 1 to 2 months group was $0.62{\pm}0.73$, in more than 2 months group was $1.75{\pm}1.03$. According to the above results, we have concluded that more than 2 months of treatment is more beneficial than 1 to 2 months of treatment.

  • PDF

회사채 신용등급 예측을 위한 SVM 앙상블학습 (Ensemble Learning with Support Vector Machines for Bond Rating)

  • 김명종
    • 지능정보연구
    • /
    • 제18권2호
    • /
    • pp.29-45
    • /
    • 2012
  • 회사채 신용등급은 투자자의 입장에서는 수익률 결정의 중요한 요소이며 기업의 입장에서는 자본비용 및 기업 가치와 관련된 중요한 재무의사결정사항으로 정교한 신용등급 예측 모형의 개발은 재무 및 회계 분야에서 오랫동안 전통적인 연구 주제가 되어왔다. 그러나, 회사채 신용등급 예측 모형의 성과와 관련된 가장 중요한 문제는 등급별 데이터의 불균형 문제이다. 예측 문제에 있어서 데이터 불균형(Data imbalance) 은 사용되는 표본이 특정 범주에 편중되었을 때 나타난다. 데이터 불균형이 심화됨에 따라 범주 사이의 분류경계영역이 왜곡되므로 분류자의 학습성과가 저하되게 된다. 본 연구에서는 데이터 불균형 문제가 존재하는 다분류 문제를 효과적으로 해결하기 위한 다분류 기하평균 부스팅 기법 (Multiclass Geometric Mean-based Boosting MGM-Boost)을 제안하고자 한다. MGM-Boost 알고리즘은 부스팅 알고리즘에 기하평균 개념을 도입한 것으로 오분류된 표본에 대한 학습을 강화할 수 있으며 불균형 분포를 보이는 각 범주의 예측정확도를 동시에 고려한 학습이 가능하다는 장점이 있다. 회사채 신용등급 예측문제를 활용하여 MGM-Boost의 성과를 검증한 결과 SVM 및 AdaBoost 기법과 비교하여 통계적으로 유의적인 성과개선 효과를 보여주었으며 데이터 불균형 하에서도 벤치마킹 모형과 비교하여 견고한 학습성과를 나타냈다.