Improved dialogue communication systems for individuals with dementia
Kazuhiro Morita, Masao Fuketa, Jun‐ichi Aoe, Kiyoshi Yasuda · International Journal of Computer Applications in Technology · 2015
With the wide spread of ubiquitous environments, many people have opportunities for using effective communication on mobile devices. Specially, cloud computing is a typical technology to achieve these desirable features efficiently. This paper presents a dialogue communication system for individuals with dementia by using cloud-based approaches. The presented method proposes a dialogue-controlling algorithm that can break communication if individuals with dementia go into a hyperexcitable state. Videos with school songs are utilised to mitigate the impacts caused by hyperexcitable state. In the cloud systems, a text mining module is combined in order to analyse personal orientation from communicated utterances of individuals with dementia. From experimental results for a total of roughly 327 individuals with dementia, it turns out that the presented method can make conversations smooth in comparison with the previous approaches. For the cloud system with text mining modules, it is verified that the presented system can achieve both high performance and low cost.