A research project funded by the National Science Foundation (NSF) develops an online tool called CitizenHelper. This tool can sort through millions of tweets to identify behaviours that could assist emergency agencies and give them an understanding of the population’s attitudes. The U.S. researchers specifically use this tool to gain insight into people’s response to COVID-19 in the Washington D.C., Maryland, and Virginia (DMV) area. The tool uses artificial intelligence (AI) techniques to filter the posts and then determine the relevance and information level of each tweet.
The head of the research team says that he and his team are extracting intelligence from social media posts on several key subjects. They include risks, symptoms, compliance with social distancing, and more relevant information on COVID-19 using a human-artificial intelligence (AI) teaming approach. The research team is also collaborating with Montgomery County, Maryland’s Community Emergency Response Team (CERT) to bring this initiative to other areas of the DMV region, including Fairfax County’s CERT.
The tool helps these CERTs to scale work that would be difficult for humans to do alone. The head of the research team says that humans are good at contextual understanding to filter content but they cannot scale. Machines, on the other hand, are good at scaling, but they do not deeply understand the context very well. Hence, a human-AI teaming approach is invaluable. The algorithms need humans to help them improve their accuracy. CitizenHelper allows this very seamless interactive mechanism for humans and computers. The humans can provide feedback to the machine on what the machine has predicted.
On their most recent run of the tool, 6,500 tweets were sampled for volunteers to look through to help identify sentiments on preventative measures for COVID-19 and risks people were taking. CERT states that about 20% of the tweets were relevant, and of that, 60% of the communications associated with risk had a negative sentiment.
Looking at social media has a huge benefit as it gives information in real-time. Therefore, it reflects people’s behaviour as opposed to expectations of what people’s behaviours are. The research team continue to work on improving the tool and teach the AI algorithms to be more specific. When a new data point comes in and the algorithm is unsure of what to do with it, a human user can provide feedback. This is a specific type of activity called active learning in the world of machine learning and AI.
A university professor and a principal investigator on the project states that the goal is to determine whether this human-AI interaction can make a community more resilient. Volunteers trained to recognise problems can better understand what is happening in the community and what is being done about it. The AI cannot learn what the important information is without being taught by humans. However, humans will always have a role because of the context associated with emergency response and how it varies by place and time.
The researchers use Twitter because its data is the most open and easiest to collect and analyse. The team could use other social media platforms such as Facebook, but that would require getting permission from every user to collect their data which is impractical.
CERT adds that CitizenHelper can automatically pull metadata and geolocation information relative to COVID prevention and risk, delivering results specific to the D.C. area and therefore more useful. Still, many emergency managers are hesitant to use such a tool.
The team has presented this project to local and federal officials who like the progress and the research, but they want the government to be more committed. Hence, the team seeks to expand their research at a national level.
