JMIR Publications

ABSTRACT

Background:

Chatbots became a promising tool to support public health initiatives during the COVID-19 pandemic. Despite their potential, little research has been done on how people interact with chatbots in a crisis. Understanding user-chatbot interactions is crucial for developing services that can respond to people’s needs during a global health emergency.

Objective:

This study examined the COVID-19 pandemic-related topics online users discussed with a commercially available chatbot and compared the sentiment expressed by users from five culturally different countries.

Methods:

We analyzed 6,594 conversation sessions related to COVID-19 from five countries between 2020 and 2021 from SimSimi, one of the world’s largest open-domain chatbots. We identified chat topics using natural language processing (NLP) methods and used the Linguistic Inquiry and Word Count (LIWC) dictionary to analyze user sentiment. Additionally, we compared the topic and sentiment variations across several English-speaking chats in the US, Malaysia, and the Philippines.

Results:

From the analysis, we extracted 18 topics, which were categorized into five themes: "Questions asked to the chatbot" (30.6%), "Preventive behaviors" (25.3%), "Outbreak of COVID-19" (16.4%), "Physical & psychological impact of COVID-19" (16.0%), and "People and life in the pandemic" (11.8%). The most common theme indicated that people considered chatbots as a source of information about the pandemic, for example, by asking health-related questions. During the lockdown, users turned to Simimi for conversation and sent emotional messages when offline social interactions became limited. Users were more likely to express negative sentiments when conversing about masks, lockdowns, case counts, and their worries about the pandemic. In contrast, positive sentiment was associated with small talk with the chatbot. We also found cultural differences; users in the US used more negative words compared to users in Asia when talking about COVID-19.

Conclusions:

This work examined user-chatbot interactions on a live platform and provides insights into people's informational and emotional needs during a global health crisis. Users sought health-related information and shared emotional messages with the chatbot, indicating the potential of future chatbots to provide accurate health information and emotional support. In the future, researchers could look into different support strategies that align with the direction of policymakers and public health officials.