 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| l |
Abundant
research on related problems
|
|
|
|
l |
DB: approximate
join, merge/purge, record linkage
|
|
|
|
l |
DL: citation
matching, author name disambiguation
|
|
|
|
l |
AI: identity
uncertainty
|
|
|
|
l |
LIS: name
authority control
|
|
|
| l |
In a nutshell,
existing approaches often do:
|
|
|
|
l |
For two
entities, e1 and e2, capture their information in data
structures,
|
|
D(e1) and D(e2)
|
|
|
|
l |
Measure the
distance or similarity between data structures: dist(D(e1),
|
|
|
D(e2)) = d
|
|
|
|
l |
Determine for
matching:
|
|
|
|
o |
If d <
threshold, then e1 and e2 are matching entities
|
|
|
| l |
Work well for
common applications
|
|
|
| l |
Ours performs
better when
|
|
|
|
l |
Entities lack useful information
|
|
