Slide 7
Slide 7 text
[Pérez-Rosas+ 2013] Pérez-Rosas, Verónica, Rada Mihalcea, and Louis-Philippe Morency. "Utterance-level multimodal sentiment analysis." Proceedings of the
51st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Vol. 1. 2013.
[Wollmer+ 2013] Wollmer, Martin, et al. "Youtube movie reviews: In, cross, and open-domain sentiment analysis in an audiovisual context." IEEE Intelligent
Systems 99.1 (2013)
[Poria+ 2015] Poria, Soujanya, Erik Cambria, and Alexander Gelbukh. "Deep convolutional neural network textual features and multiple kernel learning for
utterance-level multimodal sentiment analysis." Proceedings of the 2015 conference on empirical methods in natural language processing. 2015.
[Cambria+ 2017] Cambria, Erik, et al. "Benchmarking multimodal sentiment analysis." arXiv preprint arXiv:1707.09538 (2017).
Introduction
● マルチモーダルな感情分析
○ 複数の先行研究が存在
[Pérez-Rosas+ 2013] [Wollmer+ 2013] [Poria+ 2015]
7
✗ 先行研究では発話間の依存関係を考慮されていない
✗ SoTAモデルでは発話を独立と見なし無視 [Cambria+ 2017]
○ 動画中の発話は順序に関係がある
■ 発話単位の連続として扱う必要がある
Introduction > Related work > Method > Experiments > Conclusion