---
date: '2022-11-21T20:09:49'
hypothesis-meta:
  created: '2022-11-21T20:09:49.369906+00:00'
  document:
    title:
    - IEEEtran-7.pdf
  flagged: false
  group: __world__
  hidden: false
  id: c71vQmnYEe2a7ffZp-667A
  links:
    html: https://hypothes.is/a/c71vQmnYEe2a7ffZp-667A
    incontext: https://hyp.is/c71vQmnYEe2a7ffZp-667A/www.researchgate.net/profile/Lin-Gui-5/publication/342058196_Multi-Task_Learning_with_Mutual_Learning_for_Joint_Sentiment_Classification_and_Topic_Detection/links/5f96fd48458515b7cf9f3abd/Multi-Task-Learning-with-Mutual-Learning-for-Joint-Sentiment-Classification-and-Topic-Detection.pdf
    json: https://hypothes.is/api/annotations/c71vQmnYEe2a7ffZp-667A
  permissions:
    admin:
    - acct:ravenscroftj@hypothes.is
    delete:
    - acct:ravenscroftj@hypothes.is
    read:
    - group:__world__
    update:
    - acct:ravenscroftj@hypothes.is
  tags:
  - multi-task learning
  - topic modelling
  - NLProc
  target:
  - selector:
    - end: 8039
      start: 7778
      type: TextPositionSelector
    - exact: "e argue that mutual learningwould benefit sentiment classification since\
        \ it enriches theinformation required for the training of the sentiment clas-sifier\
        \ (e.g., when the word \u201Cincredible\u201D is used to describe\u201Cacting\u201D\
        \ or \u201Cmovie\u201D, the polarity should be positive)"
      prefix: "thewords \u201Cacting\u201D and \u201Cmovie\u201D. W"
      suffix: . At thesame time, mutual learni
      type: TextQuoteSelector
    source: https://www.researchgate.net/profile/Lin-Gui-5/publication/342058196_Multi-Task_Learning_with_Mutual_Learning_for_Joint_Sentiment_Classification_and_Topic_Detection/links/5f96fd48458515b7cf9f3abd/Multi-Task-Learning-with-Mutual-Learning-for-Joint-Sentiment-Classification-and-Topic-Detection.pdf
  text: By training a topic model that has "similar" weights to the word vector model
    the sentiment task can also be improved (as per the example "incredible" should
    be positive when used to describe "acting" or "movie" in this context
  updated: '2022-11-21T20:09:49.369906+00:00'
  uri: https://www.researchgate.net/profile/Lin-Gui-5/publication/342058196_Multi-Task_Learning_with_Mutual_Learning_for_Joint_Sentiment_Classification_and_Topic_Detection/links/5f96fd48458515b7cf9f3abd/Multi-Task-Learning-with-Mutual-Learning-for-Joint-Sentiment-Classification-and-Topic-Detection.pdf
  user: acct:ravenscroftj@hypothes.is
  user_info:
    display_name: James Ravenscroft
in-reply-to: https://www.researchgate.net/profile/Lin-Gui-5/publication/342058196_Multi-Task_Learning_with_Mutual_Learning_for_Joint_Sentiment_Classification_and_Topic_Detection/links/5f96fd48458515b7cf9f3abd/Multi-Task-Learning-with-Mutual-Learning-for-Joint-Sentiment-Classification-and-Topic-Detection.pdf
tags:
- multi-task learning
- topic modelling
- NLProc
- hypothesis
type: reply
url: /replies/2022/11/21/1669061389

---



 <blockquote>e argue that mutual learningwould benefit sentiment classification since it enriches theinformation required for the training of the sentiment clas-sifier (e.g., when the word “incredible” is used to describe“acting” or “movie”, the polarity should be positive)</blockquote>By training a topic model that has "similar" weights to the word vector model the sentiment task can also be improved (as per the example "incredible" should be positive when used to describe "acting" or "movie" in this context