Reinforcement Learning Based Personalized Neural Dialogue Generation

Saha, Tulika; Chopra, Saraansh; Saha, Sriparna; Bhattacharyya, Pushpak

doi:10.1007/978-3-030-63820-7_81

Cited by 7 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To make the VA user-adaptive, the authors in (Saha et al, 2020c(Saha et al, ,d, 2018, proposed using a sentimentbased reward function for learning a dialogue policy in a task-oriented conversation. The authors of (Saha et al, 2020a) demonstrated how reinforcement learning may be used to generate meaningful responses while training generation frameworks. In (Saha et al, 2020b(Saha et al, , 2021a, the authors show how subtleties in human communication, such as Apart from these, several other work (Wei et al, 2019;Ide and Kawahara, 2021;Huo et al, 2020) that suggests using sentiment and/or emotion as an additional input in generation frameworks either during decoding or as reward to guide the models for generating responses aligned with the user's mood or feelings.…”

Section: Related Workmentioning

confidence: 99%

A Shoulder to Cry on: Towards A Motivational Virtual Assistant for Assuaging Mental Agony

Saha¹,

Reddy²,

Das³

et al. 2022

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

Self Cite

View full text Add to dashboard Cite

Mental Health Disorders continue plaguing humans worldwide. Aggravating this situation is the severe shortage of qualified and competent mental health professionals (MHPs), which underlines the need for developing Virtual Assistants (VAs) that can assist MHPs. The data+ML for automation can come from platforms that allow visiting and posting messages in peerto-peer anonymous manner for sharing their experiences (frequently stigmatized) and seeking support. In this paper, we propose a VA that can act as the first point of contact and comfort for mental health patients. We curate a dataset, Motivational VA: MotiVAte comprising of 7k dyadic conversations collected from a peer-topeer support platform. The system employs two mechanisms: (i) Mental Illness Classification: an attention based BERT classifier that outputs the mental disorder category out of the 4 categories, viz., Major Depressive Disorder (MDD), Anxiety, Obsessive Compulsive Disorder (OCD) and Post-traumatic Stress Disorder (PTSD), based on the input ongoing dialog between the support seeker and the VA; and (ii) Mental Illness Conditioned Motivational Dialogue Generation (MI-MDG): a sentiment driven Reinforcement Learning (RL) based motivational response generator. The empirical evaluation demonstrates the system capability by way of outperforming several baselines.

show abstract