• Title/Summary/Keyword: Question Type

Search Result 535, Processing Time 0.026 seconds

A Question Type Classifier Using a Decision Tree and Lexico-syntactic Patterns (Lexico-syntactic 패턴과 결정트리를 이용한 질의 유형 분류기)

  • Kim, Hark-Soo;An, Young-Hun;Seo, Jung-Yun
    • Annual Conference on Human and Language Technology
    • /
    • 2002.10e
    • /
    • pp.189-196
    • /
    • 2002
  • 질의응답 시스템이 올바른 답변을 제시하기 위해서는 사용자의 의도를 정확하고 강건하게 파악하는 것이 매우 중요하다. 이러한 요구 사항을 만족시키기 위해서 본 논문에서는 실용적 질의응답 시스템을 위한 질의 유형 분류기를 제안한다. 제안된 질의 유형 분류기는 규칙 기반의 방법과 통계 기반의 방법을 접목시킨 하이브리드 방법을 사용한다. 제안된 방법을 사용함으로써 수동으로 규칙을 작성하는 시간을 줄일 수 있었고 정확률을 향상시킬 수 있었으며 안정성을 보장받을 수 있었다. 제안된 방법에 대한 실험에서 질의 유형을 분류하는데 86%의 정확률을 얻었다.

  • PDF

Approaches to the Analysis of Case-Control Studies of the Efficacy of Screening for Cancer

  • Weiss Noel S.;Mcknight Barbara;Stevens Nancy G.
    • 대한예방의학회:학술대회논문집
    • /
    • 1994.02b
    • /
    • pp.188-195
    • /
    • 1994
  • To an increasing extent, case-control studies are being undertaken to determine if use of early detection procedures is associated with reduced mortality from cancer. The authors recommend that in such studies the analysis focus on screening activity in cases that occurs during an interval prior to diagnosis in which the cancer is believed to be detectable and still curable and to a corresponding time period in controls. This approach places a heavy burden on the investigator to estimate accurately the period during which the tumor ought to be detectable using the test in question and to sort out reliably tests done in response to signs or symptoms of the cancer from screening tests per se. Nonetheless, the authors feel that it offers the greatest ability to discern a true benefit of screening, while minimizing the numerous potential biases that can be present in this type of study.

  • PDF

Reflectance Spectrum of Main Belt Asteroid P/2010 A2

  • Kim, Jun-Han;Ishiguro, Masateru;Hanayama, Hidekazu
    • The Bulletin of The Korean Astronomical Society
    • /
    • v.36 no.1
    • /
    • pp.41.2-41.2
    • /
    • 2011
  • P/2010 A2, a main-belt asteroid having comet-like dust trail was discovered in January 2010, but the origin of the trail is controversial. Moreno et al. (2010) reported water-ice sublimation as a mechanism for the comet-like activity, whereas other researches (Jewitt et al. 2010; Snodgrass et al. 2010) stated that impact collision contributed to the dust trail. For asteroids are categorized based on spectral shape, optical observation using different color filters makes it possible to determine the taxonomic type of P/2010 A2 nucleus, thus gives an answer to the question of activation mechanism of the object. In this presentation, we report multiband observation of P/2010 A2 in January and March 2010 with 1-meter telescope of Ishigaki-Jima Astronomical Observatory. We employed three broadband filters of g', Rc, and Ic. In this presentation, we focus on the data acquisition, the reduction and the derivation of the reflectance spectrum of debris in the trail. Finally we discuss the potential cause of dust ejection from this asteroid.

  • PDF

Risk Perceived by Consumers in Apparel Buying Situation ( I ) - Risk Types and Their Relationships with Consumers' Demographic Variables- (의복구매시 소비자가 지각하는 위험에 관한 연구( I ) -위험의 유형분류, 소비자 인구변인과의 관련을 중심으로-)

  • Kim Chanjn
    • Journal of the Korean Society of Clothing and Textiles
    • /
    • v.15 no.4 s.40
    • /
    • pp.405-416
    • /
    • 1991
  • This paper examined the risks perceived by consumers in apparel buying situation by 1) measuring the contents and perception level of risk, 2) categorizing each risk into meaningful factors (risk types), 3) analyzing the relationships between risk types and consumers' demo-graphic variables. 224 respondents deliberately selected to include each level of S demographic variables were contacted with 37-item question3.ire. Factor analysis showed that 32-item perceived risk could be categorized into 6 risk types: psychological. social, economic, time/convenience loss, fashionability loss, performance risk. Psychological risk were perceived highest in terms of perception level while social and performance risk showed relatively low perception level. 4 of 5 demographic variables including sex, education level, income, occupa-tion showed partial relationship with each risk type after ANOVA and Duncan test. Sex had the greatest influence on risk perception level and each age level (20's, 30's, 40's) showed different risk structure.

  • PDF

A Study on the Effects of Information Characteristics on the Overconfidence Phenomenon in Intuitive Probability Judgements (정보의 주요 특성이 직관적 확률판정에서의 과신현상에 미치는 영향에 관한 연구)

  • Cho, Sung-Ku
    • Journal of Korean Institute of Industrial Engineers
    • /
    • v.16 no.1
    • /
    • pp.83-98
    • /
    • 1990
  • Previous studies have shown strong tendancy toward overconfidence in intuitive probability judgements. The purpose of the present study is to investigate the relations between this overconfidence phenomenon and the three major characteristics of information, namely, the pertinance, the redundancy and the quantity. An experiment was conducted where the subjects were asked to respond to 120 questions of the same type. In each question, the subjects' task was to predict, in the light of given information, which of the two given countries would have had higher GNP in 1979 and to give the probability that their choice would be correct. The results suggests that only the pertinance of information has significant influence on the degree of overconfidence.

  • PDF

Implementation of a Question Type Classification System using Keyword Patterns (키워드 패턴을 이용한 질의유형 분류 시스템 구현)

  • Ahn, Hyeokju;Kim, MinKyoung;Kim, Harksoo
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2015.04a
    • /
    • pp.813-815
    • /
    • 2015
  • 질의응답 시스템에서 정답선택의 정확률을 향상시키기 위해 본 논문은 패턴과 휴리스틱을 기반으로 하는 질의유형 추출 시스템을 구현하는 방법을 제안한다. 질의유형은 DBPedia에서 사용하는 클래스타입을 기반으로 추출되며 질의유형에 포함하는 키워드패턴들을 수집하여 키워드패턴 데이터를 생성한다. 그 후 한국어 질의에서 많이 발생하는 유형을 분석하여 휴리스틱을 이용해 사용자가 의도한 질의 유형을 출력한다. 제안시스템은 기존 연구에 비해 구축과 수정이 쉽다는 장점이 있다.

The torsional buckling analysis for cylindrical shell with material non-homogeneity in thickness direction under impulsive loading

  • Sofiyev, A.H.
    • Structural Engineering and Mechanics
    • /
    • v.19 no.2
    • /
    • pp.231-236
    • /
    • 2005
  • This study considers the buckling of orthotropic cylindrical thin shells with material nonhomogeneity in the thickness direction, under torsion, which is a power function of time. The dynamic stability and compatibility equations are obtained first. Applying Galerkin's method then applying Ritz type variational method to these equations and taking the large values of loading parameters into consideration, analytic solutions are obtained for critical parameter values. Using those results, the effects of the periodic and power variations of Young's moduli and density, ratio of Young's moduli variations, loading parameters variations and the power of time in the torsional load expression variations are studied via pertinent computations. It is concluded that all these factors contribute to appreciable effects on the critical parameters of the problem in question.

Linguistical approach with Automatic MBTI Identification Model based on Measuring Bioelectricity Patterns

  • Hyun-Tae Kim;Ye-Jin Jin;Hye-Jin Jeon;Janghwan Kim;R. Young Chul Kim
    • International journal of advanced smart convergence
    • /
    • v.12 no.3
    • /
    • pp.200-210
    • /
    • 2023
  • Until now, it is popular to use question-and-answer-based for human personality. The current inspection of representative personality types includes Myers-Briggs Type Indicator (MBTI) and job suitability evaluations. The problem of these inspection methods is influenced by the user's environment and psychological status during MBTI inspection. To solve this problem, we proposed MBTI Identification Model based on measuring bioelectricity patterns. We adapt traditional Korean medicine, the Eight Constitution, to this model. We develop an automatic MBTI identification algorithm that maps the Eight Constitution via biological current patterns to identify MBTI personality types. By utilizing the algorithm proposed in this research, it is anticipated that users will be able to measure MBTI more easily and accurately.

Deep learning-based Answer Type Classifier Considering Topicality in Korean Question Answering (한국어 질의 응답에서의 화제성을 고려한 딥러닝 기반 정답 유형 분류기)

  • Cho, Seung Woo;Choi, DongHyun;Kim, EungGyun
    • Annual Conference on Human and Language Technology
    • /
    • 2019.10a
    • /
    • pp.103-108
    • /
    • 2019
  • 한국어 질의 응답의 입력 질문에 대한 예상 정답 유형을 단답형 또는 서술형으로 이진 분류하는 방법에 대해 서술한다. 일반적인 개체명 인식으로 확인할 수 없는 질의 주제어의 화제성을 반영하기 위하여, 검색 엔진 쿼리를 빈도수로 분석한다. 분석된 질의 주제어 정보와 함께, 정답의 범위를 제약할 수 있는 속성 표현과 육하원칙 정보를 입력 자질로 사용한다. 기존 신경망 분류 모델과 비교한 실험에서, 추가 자질을 적용한 모델이 4% 정도 향상된 분류 성능을 보이는 것을 확인할 수 있었다.

  • PDF

An Analysis of the Questions Presented in Chapters of Pattern Area in Elementary School Mathematics (초등수학의 규칙성 영역 단원에 제시된 발문의 특성 분석)

  • Do, Joowon
    • Education of Primary School Mathematics
    • /
    • v.24 no.4
    • /
    • pp.189-202
    • /
    • 2021
  • The teacher's questions presented in the problem-solving situation stimulate students' mathematical thinking and lead them to find a solution to the given problem situation. In this research, the types and functions of questions presented in chapters of Pattern area of the 2015 revised elementary school mathematics textbooks were compared and analyzed by grade cluster. Through this, it was attempted to obtain implications for teaching and learning in identifying the characteristics of questions and effectively using the questions when teaching Pattern area. As a result of this research, as grade clsuter increased, the number of questions per lesson presented in Pattern area increased. Frequency of the types of questions in textbooks was found to be high in the order of reasoning questions, factual questions, and open questions in common by grade cluster. In chapters of Pattern area, relatively many questions were presented that serve as functions to help guess, invent, and solve problems or to help mathematical reasoning in the process of finding rules. It can be inferred that these types of questions and their functions are related to the learning content by grade cluster and characteristics of grade cluster. Therefore, the results of this research can contribute to providing a reference material for devising questions when teaching Pattern area and further to the development of teaching and learning in Pattern area.