SentiWordNet은 반지도학습 기법을 바탕으로 WordNet의 Synset에 감성스코어를 매긴 어휘사전
감성스코어는
긍정(P), 부정(N), 중립(O)으로 구성되는데요 한 synset에 대해 P + N + O = 1 이 됩니다. 만약 P = 1이면 그
synset은 강렬하게 긍정적인 의미인것이고, 반대로 N = 1이면 강렬하게 부정적인 의미겠지요. P =0, N = 0이라면 O=1일테니 그
synset은 완전 중립적인 것
출처:
<https://bab2min.tistory.com/573>
Untitled
어휘사전을 받아보면
위의 표와 같은 형식으로 구성되어 있다는 것을 알 수 있습니다. http://sentiwordnet.isti.cnr.it/ 에서
CC BY-SA 라이센스에 따라 자유롭게 배포
출처:
<https://bab2min.tistory.com/573>
1985년 프린스턴
대학에서 시작된 영어 의미 어휘목록 구축 프로젝트
출처:
<https://bab2min.tistory.com/573>
영어 단어들을
같은뜻별로 묶어서 synset이라는 것을 만들고 그 synset들 간의 관계를 정리해놓은 사전입니다.
출처:
<https://bab2min.tistory.com/573>
data:image/png;base64,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
출처:
<https://bab2min.tistory.com/573>
하나의 synset은
명확한 의미를 가지게 되고, 그 의미와 관계가 있는 다른 의미들도 쉽게 찾을 수 있게 됩니다. 현재 3.1버전에서는 명사, 형용사, 동사,
부사를 대상으로 약 11만개의 synset이 구축되어 있습니다.