Putting the Con in Context: Identifying Deceptive Actors in the Game of Mafia
This work addresses the challenge of capturing contextual speaker roles in language modeling, with potential applications in game assistance, but it is incremental as it builds on existing methods in a specific domain.
The researchers tackled the problem of identifying deceptive actors in the game of Mafia by analyzing language differences between honest and deceptive roles, showing that classification models can rank deceptive players as more suspicious and that training on auxiliary tasks outperforms a standard BERT-based approach.
While neural networks demonstrate a remarkable ability to model linguistic content, capturing contextual information related to a speaker's conversational role is an open area of research. In this work, we analyze the effect of speaker role on language use through the game of Mafia, in which participants are assigned either an honest or a deceptive role. In addition to building a framework to collect a dataset of Mafia game records, we demonstrate that there are differences in the language produced by players with different roles. We confirm that classification models are able to rank deceptive players as more suspicious than honest ones based only on their use of language. Furthermore, we show that training models on two auxiliary tasks outperforms a standard BERT-based text classification approach. We also present methods for using our trained models to identify features that distinguish between player roles, which could be used to assist players during the Mafia game.