A crowdsourcing mobile game used for biomedical named-entity recognition

• Main Authors: Yen-LinHuang, 黃彥霖
• Other Authors: Tien-Hao Chang
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/74763167796516684462

碩士 === 國立成功大學 === 電機工程學系 === 103 === Biomedical articles have dramatically increased recently, driving many tools for automatically extracting valuable knowledge in natural language articles. However, to make such full automatically extracted data be considered as knowledge, a manual verification step is generally required. This manual verification, which needs to employ many domain experts, may cost more than developing exaction algorithms and is hard to last for a long time. This work aims to introduce crowdsourcing, which borrows energy from community, to solve this problem. A mobile game, Markteria was implemented to extract knowledge from natural language documents “crowdsourcingly”. In this work, named-entity recognition (NER) was chosen to be the task, since it’s a required prior step for information extraction applications. Markteria is topic independent and can be extended to other NER topic in the future.