DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Park, Jong C. | - |
dc.contributor.advisor | 박종철 | - |
dc.contributor.author | Song, Hoyun | - |
dc.date.accessioned | 2019-09-04T02:46:39Z | - |
dc.date.available | 2019-09-04T02:46:39Z | - |
dc.date.issued | 2019 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=843526&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/267041 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전산학부, 2019.2,[iv, 32 p. :] | - |
dc.description.abstract | In order to effectively diagnose depression, which is one of the most harmful mental disorders, many researchers used social media by analyzing the differences in language use. However, detecting depression from social media has problems such as a small proportion of posts with depression indicators and difficulties for distinguishing depressive symptoms from temporarily depressed feelings. To address these problems, we propose hierarchical attention with depressive indicators inspired by the process of diagnosing depression by a person with domain knowledge. Our model provides not only interpretations, but also their visualizations with learned weights through attention mechanism. With this model, we can investigate different aspects of posts with depressive indicators based on psychological theories, which will help researchers to find useful evidence for depressive characteristics. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Depression detection▼atext classification▼asocial media▼aneural network▼aattention mechanism▼avisualization | - |
dc.subject | 우울증 탐지▼a텍스트 분류▼a소셜 미디어▼a신경망 네트워크▼a어텐션 메커니즘▼a시각화 | - |
dc.title | Interpretable depression detection from social media using hierarchical attention network with depressive indicators | - |
dc.title.alternative | 계층 어텐션 네트워크와 우울증 지표 인코딩을 이용한 소셜 미디어 내 해석 가능 우울증 탐지 방법 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | 송호윤 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.