Sunday, 6 January 2019
Hall 4 (Phoenix Convention Center - West and North Buildings)
As one of the most widely used social medias in China, Weibo works similarly as Twitter in the U.S and rest of the world. Weibo is an excellent source for crowdsourcing considering China’s population. In this case, I want to explore the possibility to use Weibo data as a way of verification of historical rainfall.
The overall objective of this project is to determine how Weibo distribution will compare with historical satellite precipitation data during Typhoon Mangkhut. I collected all relevant rain related Weibos with geolocation during two day period in the middle of the Typhoon Mangkhut this past September in the Guangdong, China, use Machine Learning technology (including data abstraction, feature extraction, etc.) to process raw Weibo data, display processed Weibo data on the map, and finally compare with satellite precipitation data map from NASA.
This project should bring a new visual way to prove the value of social media in science research communities.
The overall objective of this project is to determine how Weibo distribution will compare with historical satellite precipitation data during Typhoon Mangkhut. I collected all relevant rain related Weibos with geolocation during two day period in the middle of the Typhoon Mangkhut this past September in the Guangdong, China, use Machine Learning technology (including data abstraction, feature extraction, etc.) to process raw Weibo data, display processed Weibo data on the map, and finally compare with satellite precipitation data map from NASA.
This project should bring a new visual way to prove the value of social media in science research communities.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner