2 DDQ MODULEI.E., “WHAT KIND OF *
3.2 DDQ module
i.e., “What kind of
*?”, “What year was
*held?”
The word generality score
AG
was computed using
and “Where is
*?”.
the same Mainichi newspaper text described above,
The DDQ module selects the best DQ based on its
while the SDCFG for the dependency ambiguity
linguistic appropriateness and the ambiguity of the
score
AD
for each phrase was the same as that used
phrase. The linguistic appropriateness of DQs can
in (C. Hori et. al., 2003). Eighty-two types of inter-
be measured by using a language model, N-gram.
rogative sentences were created as disambiguating
Let
Smn
be a DQ generated by inserting the
n-th
queries for each noun and noun-phrase in each ques-
phrase into the
m-th template. The DDQ module
tion and evaluated by the DDQ module. The linguis-
selects the DQ that maximizes the DQ score:
tic score
Lindicating the appropriateness of inter-
H(Smn
) =λL
L(Smn
)+λD
AD
(Pn
)+λG
AG
(Pn
),rogative sentences was calculated using 1000 ques-
tions and newspaper text extracted for three years.
where
L(·)is a linguistic score such as the loga-
The structural ambiguity score
AD
was calculated
rithm for trigram probability, and
λL
,
λD
, and
λG