Enhancing Personality Recognition in Dialogue by Data Augmentation and Heterogeneous Conversational Graph Networks
This work addresses personality recognition for enhancing human-robot interactions, but it is incremental as it builds on existing methods with specific enhancements.
The paper tackled the problem of limited speaker data and modeling dependencies in dialogue for personality recognition by introducing data augmentation via personality trait interpolation and heterogeneous conversational graph networks, achieving significant improvements over baselines on the RealPersonaChat corpus.
Personality recognition is useful for enhancing robots' ability to tailor user-adaptive responses, thus fostering rich human-robot interactions. One of the challenges in this task is a limited number of speakers in existing dialogue corpora, which hampers the development of robust, speaker-independent personality recognition models. Additionally, accurately modeling both the interdependencies among interlocutors and the intra-dependencies within the speaker in dialogues remains a significant issue. To address the first challenge, we introduce personality trait interpolation for speaker data augmentation. For the second, we propose heterogeneous conversational graph networks to independently capture both contextual influences and inherent personality traits. Evaluations on the RealPersonaChat corpus demonstrate our method's significant improvements over existing baselines.