BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deep NLP in language tutoring - Francis Bond\, Associate Professor
  at the Division of Linguistics and Multilingual Studies\, Nanyang Technol
 ogical University\, Singapore
DTSTART:20170616T140000Z
DTEND:20170616T150000Z
UID:TALK72635@talks.cam.ac.uk
CONTACT:Anita Verő
DESCRIPTION:Technology-Enabled Education is an increasingly important\npar
 t of higher education. But due to the complexity of natural\nlanguages\, t
 he available technology to teach\, drill\, and evaluate\nlanguage courses 
 is still rudimentary. Therefore second language\nlearners still depend hea
 vily on human instructors for attention and\nconstructive feedback.\n\nIn 
 this talk I show how computational resources (lexicons and\ngrammars) can 
 be used to help language learners.  In particular I\nintroduce a cross-lin
 gual vocabulary learning game\, and a new kind of\nlanguage tutoring syste
 m targeted at second language learners.  It\nmakes use of language technol
 ogy to provide tailored feedback based on\nthe student's grammatical error
 s. The system is bilingual\, using\nmachine translation (MT) technology to
  provide accurate feedback in\nthe student's first language.\n\nUngrammati
 cal sentences can be corrected in many ways\, depending on\ntheir intended
  meaning. For each ungrammatical sentence provided by a\nstudent\, the sys
 tem checks if it is well-formed using a computational\ngrammar. If there i
 s some error\, it uses special rules (mal-rules) to\nidentify both the int
 ended meaning and the error. When necessary\, it\nuses MT to ask students 
 what was meant in their first language.\nFinally\, it gives hints about th
 e errors.
LOCATION:FW11\, Computer Laboratory
END:VEVENT
END:VCALENDAR