Multilingual Persuasion Detection: Video Games as an Invaluable Data Source for NLP
This work addresses data scarcity for NLP tasks by leveraging video games, but it is incremental as it applies an existing method to a new data source.
The authors tackled the problem of multilingual persuasion detection by extracting a dataset from role-playing game dialogues and demonstrated its viability using BERT, achieving unspecified results.
Role-playing games (RPGs) have a considerable amount of text in video game dialogues. Quite often this text is semi-annotated by the game developers. In this paper, we extract a multilingual dataset of persuasive dialogue from several RPGs. We show the viability of this data in building a persuasion detection system using a natural language processing (NLP) model called BERT. We believe that video games have a lot of unused potential as a datasource for a variety of NLP tasks. The code and data described in this paper are available on Zenodo.