FluTCHA: Using fluency to distinguish humans from computers

• Main Authors: Kotaro HARA, HAJIAGHAYI, Mohammad Taghi, BENDERSON, Benjamin B.
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2015
• Subjects: Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/4012; https://ink.library.smu.edu.sg/context/sis_research/article/5014/viewcontent/flutcha_draft_2.pdf
• Institution: Singapore Management University

Improvements in image understanding technologies aremaking it possible for computers to pass traditionalCAPTCHA tests with high probability. This suggests theneed for new kinds of tasks that are easy to accomplishfor humans but remain difficult for computers. In thispaper, we introduce Fluency CAPTCHA (FluTCHA), anovel method to distinguish humans from computersusing the fact that humans are better than machines atimproving the fluency of sentences. We propose a wayto let users work on FluTCHA tests and simultaneouslycomplete useful linguistic tasks. Evaluation studiesdemonstrate the feasibility of using FluTCHA todistinguish humans from computers.