저는 딥러닝을 모릅니다 저는 딥러닝을 모릅니다

논문 리뷰

• Challenges for Toxic Comment Classification: An In-Depth Error Analysis(Betty van Aken et al., 2018)

  논문 리뷰/Error Analysis 2021. 8. 25. 20:11

  Challenges for Toxic Comment Classification: An In-Depth Error Analysis(Betty van Aken et al., 2018) ● 멀티 라벨 분류 데이터셋에 대한 아키텍처의 분류 태스크 에러 분석 중심 ● 분류 세부 결과 분석은 ensemble에 대해서만 시행 ● 사용 데이터셋 및 태스크 Wikipidia talkpages(Kaggle Toxic Comment Classifiction): 6 labels Twitter Dataset : 3 labels ● 사용 아키텍처: Logistic regression, bi-RNN(LSTM, GRU), CNN + classifier에서 발생할 수 있는 idiosyncratic wors문제와 misspell wor..

  Read More
• Metaphor Detection Using Contextual Word Embeddings From Transformers(Liu, J., O’Hara, N., Rubin, A., Draelos, R., & Rudin, C. 2020)

  논문 리뷰/Disambiguation 2021. 8. 7. 16:06

  Liu, J., O’Hara, N., Rubin, A., Draelos, R., & Rudin, C. (2020, July). Metaphor detection using contextual word embeddings from transformers. In Proceedings of the Second Workshop on Figurative Language Processing (pp. 250-255). ●구 automatic metaphor detection: hand-creafting informative features에 집중& supervised machine learning algorithm 사용 -features: POS tag, concreteness, iageabilitym semanti..

  Read More
• The Impact of Word Representations on Sequential Neural MWE Identification(Nicolas Zampieri, Carlos Ramisch, Geraldine Damnati, 2019)

  논문 리뷰/MultiWordExpression 2021. 8. 7. 15:48

  Nicolas Zampieri, Carlos Ramisch, Geraldine Damnati. The Impact of Word Representations on Sequential Neural MWE Identification. Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019), Aug 2019, Florence, Italy. pp.169 - 175, ff10.18653/v1/W19-5121f 1. finding MWEs in running text(Constant,2017) 2. PRSEME 1.1(Ramisch et al. 2018) 3. FastText(character n-gram, Bojanowski et al. 2017) 4..

  Read More
• Supervised Disambiguation of German Verbal Idioms with a BiLSTM Architecture, (Rafael Ehren et al., 2020)

  논문 리뷰/MultiWordExpression 2021. 8. 7. 15:44

  ★Supervised Disambiguation of German Verbal Idioms with a BiLSTM Architecture (Rafael Ehren, Timm Lichte, Laura Kallmyer, Jakub Waszczuk, 2020) 연구방법 1. 말뭉치 ● Verbal Idiom을 위한 말뭉치 구축: COLF-VID -3명의 annotator가 idiom/literal/undeciable/both 태깅 -2명의 annotator가 context에 대해서도 태깅 2. 실험 architecture ● 워드 임베딩을 얻기 위한 모델: Word2Vec(Skip-gram), FastText(CBOW), ELMo -세 모델은 모두 pre-trained, 각자 사전 훈련된 말뭉치가 다름 -F..

  Read More