Time-Dependent
Similarity Measure
Similarity Measure
nIn relating Q2 to relevant HLFi1, how to reduce the noise
introduced by the dictionaries, esp, the gloss?
nSeveral
sources of noise:
nTypical examples
n{Story 1: Forest fire},
{Story 2: Explosion and
bombing},
{Story 3: Bombing and Fire}
nFire à
found in stories 1, 2, 3
nExplosion à
found in stories 2, 3 (but 1 may be retrieved as fire is closely related to explosion)
n{car, boat, aircraft}
nRelated via modes of transportation
(similar nature)
nBut we cannot use the concept car to
find concept boat…