移至主內容

Communal Data Workflow in TaiRON (Taiwan Roadkill Observation Network)

發表日期
2016/9/11~13
發佈資訊
SciData2016Conference
Tyng-Ruey Chuang 1, Te-En Lin 2, Yi-Hong Chang 3, Chih-Yun Chen 2, Yu-Kai Chen 2,
Ping-Keng Hsieh 1,2, Guan-Shuo Mai 4
1) Institute of Information Science, Academia Sinica, Taiwan.
2) Taiwan Endemic Species Research Institute, Council of Agriculture, Taiwan.
3) Word Gleaner Ltd., Taiwan.
4) Biodiversity Research Center, Academia Sinica, Taiwan.
授權條款
創用CC-姓名標示
姓名標示
Tyng-Ruey Chuang

Summary
TaiRON (Taiwan Roadkill Observation Network) is a collaborative project in collecting records about road kills and other types of animal mortality incidents in Taiwan. It first started in a social media setting – a Facebook group – where observations of road kills (including images, locations, dates, and other data) were submitted by volunteers. Experts identified species in the observation records, and used the collected records for research. As of 2016-07-10, there are 10,674 volunteers in the Facebook group. The data collection process in TaiRON has been communal in nature – without the volunteers, there will be no data. Furthermore, the data so collected shall be useable by the volunteers as well. Before sharing the data, however, there are a number of issues to address: the quality and period of data releases, the handling of sensitive data (the locations of endangered species, for example) and personal privacy, as well as dataset formats and access APIs. The data acquisition and management workflow in TaiRON has undergone many changes since the project started in 2011. Mobile phone apps are now used for reporting, and the observations are first uploaded to the TaiRON website before they are announced to the Facebook group. User feedbacks, such as species identifications, from the Facebook groups are incorporated in order to make high-quality datasets. In this presentation we will summarize the current data workflow in TaiRON, emphasizing its communal characteristics. 
communal_data_workflow_in_tairon.pdf
http://www.scidatacon.org/2016/sessions/46/paper/129/