• Title/Summary/Keyword: 해소

Search Result 4,491, Processing Time 0.036 seconds

Embedding with different levels for idiom disambiguation (관용표현 중의성 해소를 위한 다층위 임베딩 연구)

  • Park, Seo-Yoon;Kang, Ye-Jee;Kang, Hye-Rin;Jang, Yeon-Ji;Kim, Han-Saem
    • Annual Conference on Human and Language Technology
    • /
    • 2021.10a
    • /
    • pp.167-172
    • /
    • 2021
  • 관용표현 중에는 중의성을 가진 표현이 많다. 즉 하나의 표현이 맥락에 따라 일반적 의미와 관용적 의미 두 가지 이상으로 해석될 가능성이 있어 이런 유형의 관용표현을 중의성 해소 없이 자연어 처리 태스크에 적용할 경우 문제가 발생하게 된다. 본 연구에서는 관용표현의 특성인 중의성과 더불어 '관용표현은 이미 사용자의 머릿속에 하나의 토큰으로 저장되어 있다'라는 'Idiom Principle'을 바탕으로 관용표현에 대해 각각 표면형, 단순 단일 토큰형, stemming 단일 토큰형 층위의 임베딩을 만들어 관용표현 분류 연구를 진행하였으며, 실험 결과 표면형 및 stemming을 적용하지 않은 단순 단일 토큰으로 학습하는 것보다, stemming을 적용한 후 단일 토큰으로 학습하는 것이 관용표현의 중의성 해소에 유의미한 효과가 있음을 확인하였다.

  • PDF

Word Sense Disambiguation Using Knowledge Embedding (지식 임베딩 심층학습을 이용한 단어 의미 중의성 해소)

  • Oh, Dongsuk;Yang, Kisu;Kim, Kuekyeng;Whang, Taesun;Lim, Heuiseok
    • Annual Conference on Human and Language Technology
    • /
    • 2019.10a
    • /
    • pp.272-275
    • /
    • 2019
  • 단어 중의성 해소 방법은 지식 정보를 활용하여 문제를 해결하는 지식 기반 방법과 각종 기계학습 모델을 이용하여 문제를 해결하는 지도학습 방법이 있다. 지도학습 방법은 높은 성능을 보이지만 대량의 정제된 학습 데이터가 필요하다. 반대로 지식 기반 방법은 대량의 정제된 학습데이터는 필요없지만 높은 성능을 기대할수 없다. 최근에는 이러한 문제를 보완하기 위해 지식내에 있는 정보와 정제된 학습데이터를 기계학습 모델에 학습하여 단어 중의성 해소 방법을 해결하고 있다. 가장 많이 활용하고 있는 지식 정보는 상위어(Hypernym)와 하위어(Hyponym), 동의어(Synonym)가 가지는 의미설명(Gloss)정보이다. 이 정보의 표상을 기존의 문장의 표상과 같이 활용하여 중의성 단어가 가지는 의미를 파악한다. 하지만 정확한 문장의 표상을 얻기 위해서는 단어의 표상을 잘 만들어줘야 하는데 기존의 방법론들은 모두 문장내의 문맥정보만을 파악하여 표현하였기 때문에 정확한 의미를 반영하는데 한계가 있었다. 본 논문에서는 의미정보와 문맥정보를 담은 단어의 표상정보를 만들기 위해 구문정보, 의미관계 그래프정보를 GCN(Graph Convolutional Network)를 활용하여 임베딩을 표현하였고, 기존의 모델에 반영하여 문맥정보만을 활용한 단어 표상보다 높은 성능을 보였다.

  • PDF

Dam-Operation Limited Subject Dissolution of Multi-Dam Down Stream River and Recoverly Effect Analysis of Flood Control Funtion (다목적댐 하류하천의 댐운영 제약사항 해소와 치수기능 회복효과 분석)

  • Yang, Seung-In;Ryu, Tae-Sang
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2009.05a
    • /
    • pp.240-244
    • /
    • 2009
  • 다목적댐 하류하천은 댐 건설당시 정비가 필요한 지역으로 댐 방류의 영향을 받는 일정구간에 대해서는 댐설계방류량에 대응하도록 하천정비가 시행되어야 하나, 지금까지의 댐 하류하천은 댐과 별개로 하천정비가 시행되어 왔으며, 관리 또한 댐과 이원화되어 이루어져 왔다. 이로 인해 댐하류하천은 제방월류, 하천구역내 사유지 및 불법시설물 등의 댐운영 제약사항으로 댐 본래의 치수기능이 저하되어 한국수자원공사에서 관리중인 15개 다목적댐 여수로 방류가 설계방류량의 16%(15개 다목적댐 전체 여수로 설계방류량 $57,004m^3/s$ ${\rightarrow}$ 무피해방류량 $9,300m^3/s$)에 불과하고, 대규모 홍수 발생시 댐하류지역에서 막대한 피해와 민원을 발생시키고 있다. 이러한 댐운영 제약사항해소를 위해 한국수자원공사에서는 '07년부터 연차별로 소양강댐을 포함한 22개댐 하류하천의 정비를 시행하고 있다. 현재 시행중인 용담댐 하류하천정비를 시행사례를 통해 댐운영 제약사항해소와 이에 따른 댐의 치수기능 회복효과를 분석해보면 다음과 같다. 댐방류시 현장조사를 통해 확인된 댐운영 제약사항은 하천구역내 사유지를 포함해하여 총 8건이며, 구조적 대책인 제방 축조와 비구조적 대책인 사유지 보상 등을 통해 댐 무피해방류량을 현재 $700m^3/s$ (여수로 설계방류량 대비 22%)에서 $2,730m^3/s$(여수로 설계방류량 대비 85%)로 증가 방류 할 수있는 것으로 분석되었다. 그러나, 구조적 비구조적 대책으로 불가능한 하천구역내 불법 경작지 및 시설물 등이 일부 잔존하여 댐운영 제약사항으로 여전히 남아 있는 실정이다. 이러한 문제를 해소하기 위해서는 불법행위에 대한 관리감독 강화 등의 조치가 필요하나, 이원화된 현행 관리체계에서는 한계가 있으며, 이와 같은 문제를 해소하기 위해서는 댐 방류의 영향을 받는 하류하천 일정구간에 대해서는 댐관리자의 역할 강화가 필요하다.

  • PDF

A Study on Statistical Feature Selection with Supervised Learning for Word Sense Disambiguation (단어 중의성 해소를 위한 지도학습 방법의 통계적 자질선정에 관한 연구)

  • Lee, Yong-Gu
    • Journal of the Korean BIBLIA Society for library and Information Science
    • /
    • v.22 no.2
    • /
    • pp.5-25
    • /
    • 2011
  • This study aims to identify the most effective statistical feature selecting method and context window size for word sense disambiguation using supervised methods. In this study, features were selected by four different methods: information gain, document frequency, chi-square, and relevancy. The result of weight comparison showed that identifying the most appropriate features could improve word sense disambiguation performance. Information gain was the highest. SVM classifier was not affected by feature selection and showed better performance in a larger feature set and context size. Naive Bayes classifier was the best performance on 10 percent of feature set size. kNN classifier on under 10 percent of feature set size. When feature selection methods are applied to word sense disambiguation, combinations of a small set of features and larger context window size, or a large set of features and small context windows size can make best performance improvements.

A Study on Considerations of the Library Policy Establishment to Resolve Digital Divide of Older Adults (노인의 정보격차 해소를 위한 도서관 정책수립의 고려사항)

  • Yoon, Hye-Young
    • Journal of Korean Library and Information Science Society
    • /
    • v.48 no.4
    • /
    • pp.77-105
    • /
    • 2017
  • As our society enters an aged society, it is becoming increasingly important to resolve digital divide of older adults. However, the digital divide of the older adults are generally neglected compared to other information poor. The aim of this study is to propose considerations for library policy development in order to resolve the digital divide of the older adults. The study examined regulation of the Framework Act on National Informatization and the Ministry of Science and ICT, the Ministry of Culture, Sports and Tourism, the Ministry of Health and Welfare, the Ministry of Employment and Labor. Also it investigated guidelines and current status of services to older adults in the library. The study suggested that public libraries should be used as a basic base for the services to resolve the digital divide to older adults across the country, and the services should be organized systematically and sustainably through public libraries.

Effects of Slander Glasswort (Salicornia herbacea L.) Extract on Improvements in Bowel Function and Constipation Relief (함초 추출물의 장 기능 개선과 변비 해소 효과)

  • Cho, Young-Sim;;Han, Young-Sil
    • Korean Journal of Food Science and Technology
    • /
    • v.40 no.3
    • /
    • pp.326-331
    • /
    • 2008
  • This study investigated the effects of dietary slander glasswort extract on improvements in bowel function and constipation relief. Balb/C mice were used to measure transit movement of a charcoal meal after slander glasswort extract was added to their drinking water at 0.36, 0.72, and 1.5 mg/mL concentrations. The extract's constipation relief effects were investigated in Sprague-Dawley rats with loperamide-induced constipation by adding 2, 3, and 4 mg/mL of the extract to their drinking water, and by feeding yogurt containing 0.0, 0.25, and 0.50% slander glasswort extract mixed into a 10-fold amount of water. The mice receiving the extract at 1.5 mg/mL via drinking water had a 16.4% increase in the rate of transit as compared to the controls. In addition, the rats receiving the extract via drinking water and yogurt had increased in constipation relief from 128 to 156% and from 107 to 150%, respectively, as compared to the controls. Based on the results, slander glasswort is considered to have effects on improving bowel function and relieving constipation.

Discrepancy between Reading and Writing Equality Number Sentences in Korean Language (등호 해석의 두 시간적 차원인 읽기.쓰기의 불일치와 그 해소)

  • Yim, Jaehoon
    • Journal of Elementary Mathematics Education in Korea
    • /
    • v.17 no.2
    • /
    • pp.207-223
    • /
    • 2013
  • Teachers unfold a series of timeless mathematical symbols such as 5+2=7 in time by verbalizing the symbols in classrooms. A number sentence 5+2=7 is read in Korean as '5 더하기 2는(five plus two) 7과(seven) 같다(equals). Unlike in English, 5+2 and 7 are read first before the equal sign in Korean. This sequence of reading in Korean conflicts with the conventional linguistic sequence of writing from left to right. Ways of resolving the discrepancy between reading and writing sequences can make a difference students' understanding of the equal sign. Students would be in danger of perceiving the equal sign as an operational symbol, if a teacher resolves the discrepancy by subordinating reading sequence to linguistic convention of writing. This way of resolving results in the undesired phenomenon of changing the reading expressions in Korean elementary math textbook which represent relational notion of the equal sign into other reading expressions that represent operational notion of it. For understanding of relational notion of the equal sign, the discrepancy should be resolved by changing writing sequence in accordance with reading sequence. In addition, teaching of verbalizing the equal sign should be integrated with teaching of verbalizing inequality signs.

  • PDF

A Study on Bridging the Digital Divide of K-12 Students of Multicultural Background (다문화가정 초중등학생을 위한 정보격차해소 방안 연구)

  • Lee, June
    • The Journal of Korean Association of Computer Education
    • /
    • v.12 no.6
    • /
    • pp.41-51
    • /
    • 2009
  • The purpose of this study was to suggest policy implications for bridging digital divide of the students of multicultural background. Specifically, through maximizing equity of Information technology, the current study suggested the need for greater intervention and comprehensive strategies in the interests of a socially inclusive information society, of benefit to the IT disadvantaged. To achieve the research goal, I reviewed a variety of policy reports and previous studies related to the issue of digital divide and multicultural education, and draw useful implications for the policy change. The major research results were as the following: a policy change in a support of personal computer toward public place rather than individual home, widening use of e-learning service as a tool for bridging the digital divide, and the training for teachers and parents to provide an effective education to reduce the digital divide.

  • PDF

A Study on an Algorithm Based on Sensitivity Method for Alleviating Overloads in Power Networks (송전선로 과부하를 해소하기 위한 민감도에 근거한 알고리즘 연구)

  • Lee, Byung-Ha;Baek, Jung-Myoung
    • Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
    • /
    • v.22 no.8
    • /
    • pp.96-103
    • /
    • 2008
  • In this paper, a new algorithm based on sensitivity method for alleviating overloads in power networks is presented to find the switching branches effectively. By applying the new sensitivity of the line flow with respect to the change of the branch impedance, both on and off switchings for alleviating overloads in power networks are performed systematically at once and an effective scheme for drawing up a plan to alleviate overloads of the lines is presented through screening a large number of line switching cases easily. The ranking of switching branches is calculated according to the new switching algorithm based on sensitivity method and the switching of the ranked branches is performed in the order of ranking until overloads are eliminated. In order to show the effects of this algorithm, it is applied to a small scale power system of IEEE 39-bus test system.

Word Sense Disambiguation of Predicate using Sejong Electronic Dictionary and KorLex (세종 전자사전과 한국어 어휘의미망을 이용한 용언의 어의 중의성 해소)

  • Kang, Sangwook;Kim, Minho;Kwon, Hyuk-chul;Jeon, SungKyu;Oh, Juhyun
    • KIISE Transactions on Computing Practices
    • /
    • v.21 no.7
    • /
    • pp.500-505
    • /
    • 2015
  • The Sejong Electronic(machine readable) Dictionary, which was developed by the 21 century Sejong Plan, contains a systematic of immanence information of Korean words. It helps in solving the problem of electronical presentation of a general text dictionary commonly used. Word sense disambiguation problems can also be solved using the specific information available in the Sejong Electronic Dictionary. However, the Sejong Electronic Dictionary has a limitation of suggesting structure of sentences and selection-restricted nouns. In this paper, we discuss limitations of word sense disambiguation by using subcategorization information as suggested by the Sejong Electronic Dictionary and generalize selection-restricted noun of argument using Korean Lexico-semantic network.