Estimation of Emotion Type and Intensity in Japanese Tweets Using Multi-task Deep Learning

2Citations
Citations of this article
7Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Accurate estimation of emotions in SNS posts plays an essential role in a wide variety of real world applications such as intelligent dialogue systems, review analysis for recommendations and so on. In this paper, we focus on developing accurate models for estimating types of emotions and their intensities in Japanese tweets by using multi-task deep learning. More concretely, three deep learning models for estimating intensities of emotions were extended to be able to predict the emotional types and their intensities at a time. The effectiveness of the developed models was confirmed through experiments using the database of Japanese tweets annotated with intensity scores of four types of emotions using best-worst scaling.

Cite

CITATION STYLE

APA

Sato, K., & Ozaki, T. (2019). Estimation of Emotion Type and Intensity in Japanese Tweets Using Multi-task Deep Learning. In Advances in Intelligent Systems and Computing (Vol. 927, pp. 314–323). Springer Verlag. https://doi.org/10.1007/978-3-030-15035-8_30

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free