BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Ke Deng/邓柯 (Center for Statistical Science\, Tsinghua Universi
 ty)
DTSTART:20201227T083000Z
DTEND:20201227T091500Z
DTSTAMP:20260423T041339Z
UID:iccm2020/18
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/iccm2020/18/
 ">Understanding Chinese Texts via Statistical Inference</a>\nby Ke Deng/
 邓柯 (Center for Statistical Science\, Tsinghua University) as part of I
 CCM 2020\n\n\nAbstract\nWith the growing availability of digitized text da
 ta both publicly and privately\, there is a great need for effective compu
 tational tools to automatically extract information from texts. Because th
 e Chinese language differs most significantly from alphabet-based language
 s in not specifying word boundaries\, most existing Chinese text-mining me
 thods require a prespecified vocabulary and/or a large relevant training c
 orpus\, which may not be available in some applications. We proposed a fam
 ily of statistical approaches that can achieve multiple NLP tasks\, such a
 s word discovery\, name entity recognition\, word segementation\, semamtic
  understanding and relation extraction\, simultaneously with little traini
 ng information. These approaches are particularly useful for mining domain
 -specific texts where the underlying vocabulary is unknown and/or the text
 s of interest differ significantly from standard training corpora.\n
LOCATION:https://researchseminars.org/talk/iccm2020/18/
END:VEVENT
END:VCALENDAR