A Cognitive Stimulation Dialogue System with Multi-source Knowledge Fusion for Elders with Cognitive Impairment

Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 10628–10640 July 9-14, 2023 ©2023 Association for Computational Linguistics A Cognitive Stimulation Dialogue System with Multi-source Knowledge Fusion for Elders with Cognitive Impairment Jiyue Jiang, Sheng Wang, Qintong Li, Lingpeng Kong, Chuan Wu The University of Hong Kong {jiangjy,u3009618,qtli}@connect.hku.hk {lpk,cwu}@cs.hku.hk Abstract When communicating with elders with cog- nitive impairment, cognitive stimulation (CS) help to maintain the cognitive health of elders. Data sparsity is the main challenge in building CS-based dialogue systems, particularly in the Chinese language. To fill this gap, we construct a Chinese CS conversation (CSConv) dataset, which contains about 2.6K groups of dialogues with therapy principles and emotional support strategy labels. Making chit chat while pro- viding emotional support is overlooked by the majority of existing cognitive dialogue systems. In this paper, we propose a multi-source knowl- edge fusion method for CS dialogue (CSD), to generate open-ended responses guided by the therapy principle and emotional support strat- egy. We first use a progressive mask method based on external knowledge to learn encoders as effective classifiers, which is the prerequisite to predict the therapy principle and emotional support strategy of the target response. Then a decoder interacts with the perceived therapy principle and emotional support strategy to gen- erate responses. Extensive experiments con- ducted on the CSConv dataset demonstrate the effectiveness of the proposed method, while there is still a large space for improvement com- pared to human performance1. 1 Introduction Dialogue systems have enjoyed rapid progress in re- cent years, through communication with humans to satisfy diverse needs (Liu et al., 2021; Kann et al., 2022). Cognition stimulation of elders is a criti- cal psychological therapy where dialogue systems serve as effective tools for restoring the cognition of older adults (De Oliveira et al., 2014; Park et al., 2019; Tokunaga et al., 2021). Some studies have shown that chit-chat can help older people with cognitive restoration (van Rijn 1Our data and code could be found in https://github. com/jiangjyjy/CSD I remember the time when I used to work in Hong Kong. (None, Reminiscence, Self-disclosure) I should understand his/her situation. Let me ask him/her about his time working in Hong Kong. Did you enjoy the time being there? (None, Inquiry, Question) Not really. I was a waiter and life was hard at that time. (Sadness, Reminiscence, Self-disclosure) I should comfort his/her and ask her more questions to stimulate his/her mind and expression. Oh, I understand you. I thought you were a teacher then! Anyway, do you think teacher is better compared with waiter? (Surprise, Reminiscence , Question) I cannot decide. I guess teacher is better. (None, Expression , Self- disclosure) I should ask her why to encourage him/her to think further. Why so? Tell me more about it. (None, Inquiry, Question) Figure 1: An example of a Chinese CS-based dialogue from the CSConv dataset (translated from Chinese to En- glish), being provided to the elders (left) by the therapist (right). The emotion classification, …

Content truncated to 500 words