Location

Hilton Waikoloa Village, Hawaii

Event Website

http://www.hicss.hawaii.edu

Start Date

1-4-2017

End Date

1-7-2017

Description

This action research adopts a case study approach of reliability assurance and need finding for radiation monitoring on social media communication through the development and use of Pocket Geiger (POKEGA), a smartphone-connected radiation detector developed after the Fukushima Nuclear Disaster in Japan. We also facilitated the use of an inclusive Facebook community for radiation monitoring established by volunteer experts and normal users. Interaction through this social media led to credible discussions and enabled dose verification among users. It is important to grasp, at least roughly, the potential needs of and/or information required by users, especially those living in and around affected areas, from among the numerous topics posted by regular citizens and radiation experts on social media. Therefore, we developed a Web service called Crowd Talks to summarize and visualize discussions within the POKEGA community. This tool provides quick analysis and visualization of discussion topics from POKEGA Facebook posts based on natural language processing algorithms such as MeCab, latent Dirichlet allocation, and principal component analysis. Our results suggested that Crowd Talks was able to identify major topic clusters from the Facebook community.

Share

COinS
 
Jan 4th, 12:00 AM Jan 7th, 12:00 AM

Wisdom of Crowds for Reliable Discussion and Need Finding: A Case Study of Information Sharing Regarding Radiation after the Fukushima Nuclear Disaster

Hilton Waikoloa Village, Hawaii

This action research adopts a case study approach of reliability assurance and need finding for radiation monitoring on social media communication through the development and use of Pocket Geiger (POKEGA), a smartphone-connected radiation detector developed after the Fukushima Nuclear Disaster in Japan. We also facilitated the use of an inclusive Facebook community for radiation monitoring established by volunteer experts and normal users. Interaction through this social media led to credible discussions and enabled dose verification among users. It is important to grasp, at least roughly, the potential needs of and/or information required by users, especially those living in and around affected areas, from among the numerous topics posted by regular citizens and radiation experts on social media. Therefore, we developed a Web service called Crowd Talks to summarize and visualize discussions within the POKEGA community. This tool provides quick analysis and visualization of discussion topics from POKEGA Facebook posts based on natural language processing algorithms such as MeCab, latent Dirichlet allocation, and principal component analysis. Our results suggested that Crowd Talks was able to identify major topic clusters from the Facebook community.

http://aisel.aisnet.org/hicss-50/cl/crisis_and_disaster_management/11