Multi-task dialog act and sentiment recognition on Mastodon

TL;DR

This paper introduces a new annotated Mastodon corpus for dialog act and sentiment recognition, along with a multi-task neural network, addressing reproducibility issues in social media research.

Contribution

It provides a licensed, annotated Mastodon dataset and a multi-task hierarchical model for joint dialog act and sentiment recognition, enabling reproducible research.

Findings

Transfer learning is effective between dialog act and sentiment tasks.

Correlations between sentiments and dialogue acts are identified.

Open-source dataset and model facilitate future research.

Abstract

Because of license restrictions, it often becomes impossible to strictly reproduce most research results on Twitter data already a few months after the creation of the corpus. This situation worsened gradually as time passes and tweets become inaccessible. This is a critical issue for reproducible and accountable research on social media. We partly solve this challenge by annotating a new Twitter-like corpus from an alternative large social medium with licenses that are compatible with reproducible experiments: Mastodon. We manually annotate both dialogues and sentiments on this corpus, and train a multi-task hierarchical recurrent network on joint sentiment and dialog act recognition. We experimentally demonstrate that transfer learning may be efficiently achieved between both tasks, and further analyze some specific correlations between sentiments and dialogues on social media. Both the annotated corpus and deep network are released with an open-source license.