• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 課題ページに戻る

2018 年度 実施状況報告書

大規模の逆強化学習ための文脈データを収集するシステムの開発

研究課題

研究課題/領域番号 17K00295
研究機関北海道大学

研究代表者

RZEPKA Rafal  北海道大学, 情報科学研究科, 助教 (80396316)

研究期間 (年度) 2017-04-01 – 2021-03-31
キーワードcontext processing / implicit knowledge / semantic chains / text generation
研究実績の概要

The second year of the grant realization was concentrated on a) developing methods for the new knowledge generation in order to allow machines to read context which is not explicitly shown in Japanese text b) testing automatically retrieved knowledge and contexts with machine learning methods on English (motivation discovery) and Chinese (sentiment analysis).
a) In the past, researchers calculated the relevance between two events appearing in the same document using distributed representations for machine learning. Since such methods handle only explicitly written information, it is difficult for a machine to, e.g. automatically answer questions about context. To tackle this problem, a method for generating event chains using relevance filtering and for augmenting implicit knowledge as places as an example was created. Temporal events were created and their implicit relationship by relevance using a distributed representation of words was calculated. Likewise, relevant places were also added. Experiments were performed to show the temporal and semantic naturalness of the knowledge completion process needed to enrich machine learning.
b) Various experiments on the influence of context were performed and their results were presented during national and international conferences. The main findings were that 1) emoticons are one of the means which can enrich context and provide additional information helping to estimate affect beyond positivity and negativity; 2) that computer can learn a motivation degree of an advice depending on the context of the request for this advice.

現在までの達成度 (区分)
現在までの達成度 (区分)

2: おおむね順調に進展している

理由

I have discovered that existing corpora give a vast source of knowledge, however, the machines lack an ability to fill in missing parts of context as places and agents, which is well visible in Japanese language. For this reason, I needed to work on a new algorithm to deal with this phenomenon which slightly delayed the process of generating Shankian scripts. Nevertheless, planned work on contextual understanding of emotions and motivations generated more results as expected, therefore I assess this year progress as acceptedly paced in general.

今後の研究の推進方策

I will continue developing the knowledge acquisition system that adds implicit knowledge about given situations. As stated in the research plan, I will broaden tests of the acquired knowledge beyond the sentiment and experiment with vectors for deeper semantic representations to discover further dependencies between places, actors and acts to allow understanding short utterances by extending assumptions about the context. I plan to show how, through interaction with users, a machine can narrow the enormous data of possible interpretations. This approach should lead to higher explainability of AI systems, especially in the automatic moral judgement scenarios.

  • 研究成果

    (9件)

すべて 2019 2018

すべて 学会発表 (9件) (うち国際学会 3件)

  • [学会発表] 単語の分散表現を用いた日本語イベント連鎖の自動構築2019

    • 著者名/発表者名
      瀧下祥, Rafal Rzepka, 荒木健治
    • 学会等名
      言語処理学会第25回年次大会(NLP2019)
  • [学会発表] Unicorn Story Generation and Limits of Words - On Perspectives of Automatic Tacit Knowledge Addition2019

    • 著者名/発表者名
      Rafal Rzepka, Sho Takishita and Kenji Araki
    • 学会等名
      JSAI Special Interest Group for Artificial General Intelligence
  • [学会発表] Analyzing motivating texts for modelling human-like motivation techniques in emotionally intelligent dialogue systems2018

    • 著者名/発表者名
      Patrycja Swieczkowska, Rafal Rzepka and Kenji Araki
    • 学会等名
      Biologically Inspired Cognitive Architectures (BICA 2018) Proceedings, Springer Series on Advances in Intelligent Systems and Computing
    • 国際学会
  • [学会発表] Emoticon-Aware Recurrent Neural Network Model for Chinese Sentiment Analysis2018

    • 著者名/発表者名
      Da Li, Rafal Rzepka, Michal Ptaszynski and Kenji Araki
    • 学会等名
      Proceedings of The Ninth IEEE International Conference on Awareness Science and Technology
    • 国際学会
  • [学会発表] Machine Learning Approach Considering Chinese Slang Lexicon and Emoticons for Chinese Social Media Sentiment Analysis2018

    • 著者名/発表者名
      Da Li, Rafal Rzepka, Michal Ptaszynski, Kenji Araki
    • 学会等名
      AAAI-19 Workshop on Affective Content Analysis
    • 国際学会
  • [学会発表] First Trials with Culture-Dependent Moral Commonsense Acquisition2018

    • 著者名/発表者名
      Rafal Rzepka, Da Li and Kenji Araki
    • 学会等名
      The 32th Annual Conference of the Japanese Society for Artificial Intelligence
  • [学会発表] Preliminary Analysis of Weibo Emojis for Sentiment Analysis of Chinese Social Media2018

    • 著者名/発表者名
      Da Li, Rafal Rzepka and Kenji Araki
    • 学会等名
      32th Annual Conference of the Japanese Society for Artificial Intelligence
  • [学会発表] Retrieving Metaphorical Sentences from Japanese Literature Using Standard Text Classification Methods2018

    • 著者名/発表者名
      Mateusz Babieno, Sho Takishita, Rafal Rzepka and Kenji Araki
    • 学会等名
      人工知能学会第2種研究会 ことば工学研究会
  • [学会発表] 常識的知識の自動評価手法及び英日辞書を用いたConceptNetの拡張2018

    • 著者名/発表者名
      首藤聖矢,ジェプカ・ラファウ,荒木健治
    • 学会等名
      ARG Web インテリジェンスとインタラクション研究会第13回研究会

URL: 

公開日: 2019-12-27  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi