Welcome to Journal of Graphics share: 

Journal of Graphics

Previous Articles     Next Articles

Proactive interaction design of conversational agent for smart homes

  

  1. (1. School of Design, Hunan University, Changsha Hunan 410000, China;
    2. Baidu Online Network Technology (Beijing) Co. Ltd, Beijing 100089, China)
  • Online:2020-08-31 Published:2020-08-22

Abstract: Aimed at the proactive interaction design strategies of conversational agents (CAs) in the
smart home, the proactivity was discussed in this article based on decision right and communication
style. In this light, four proactive styles (straight forward decision-maker and adviser, euphemistic
decision-maker and adviser) were designed for smart home CAs. A Wizard-of-OZ experiment was
conducted to investigate the effects of users’ gender, age, experience and living status on their
preferences for proactive styles. The results reveal that: ① Users having had experience with smart
home products are more satisfied with the “adviser” style; family users feel more comfortable with
the “adviser” style. ② Users generally feel more satisfied and comfortable when the CA
communicates with them in an explicit manner. And the experiment verifies that: ① Users’ prior
experience with smart home products and their living states have an impact on users’ preferences for
proactive styles of smart home CAs. ② In the smart home scenario, it is found that the
communication efficiency of CAs is more important than politeness.

Key words: proactive interaction, human-computer interaction, conversational agents, smart homes;
user characteristics