August 17, 2005
Generic Soft Pattern Models for Definitional QA
14/28
Bigram Model in Dealing with Gaps
•Bigram model can deal with gaps
–Unseen tokens have small smoothing probabilities in specific positions
•
<Search_Term>  which   is   known   for    DT$    NNP
Test sentence:
Pattern
<Search_Term>      ,    whose   book  is   known   for
P(“,”|S1) = P(“whose”|S2) = P(“book”|S3)  = P(“is”|S4) =
P(“,”|S1) = P(“whose”|S2) … = small smoothing prob
P(“known”|S3) = 0.3      P(“for”|S4) = 0.21
Not too good!