Local context

1
A confidence-based framework for disambiguating
geographic terms Erik Rauch, Michael Bukatin,
and Kenneth Baker MetaCarta, Inc.
2
(No Transcript)
3
wine in Europe
4
Al Hamra
( red in Arabic)
5
(No Transcript)
6
Local and non-local information
More non-local information -gt too many states to
get probabilities
Madison
s downtown
Wisconsin
Milwaukee
7
Candidate places
8
Local context
resident of Madison
Madison, WI Madison, ID Madison, CT Madison,
KY
9
Context affects confidence

• Increase or decrease c(p,n) based on strength of
  context words
• by Madison vs. President Madison
• can be added manually or automatically
• and/or use HMM

10
Local context problems
Madison family attractions
Milwaukee
Madison, WI Madison, ID Madison, CT Madison,
KY
11
Using spatial patterns of geographic references
12

Increase c(p,n) based on number of other
references Enclosing regions or nearby points
Madison
Wisconsin
Milwaukee
13
Pitfalls
14
Training

• Philadelphia is usually geographic Bend
  usually isnt
• If name n often refers to point p in documents,
  give (n,p) high confidence to start with
• Use average confidence in a large corpus

15
Training contd

• Extract local linguistic contexts that often
  occur with geographic names in tagged corpora
• Or train HMM

16
Relevance
Query cheese in France

• Several dimensions to relevance
• Traditional textual relevance of query terms
• Georelevance

17
Georelevance

• Depends on
• Attributes of the geotext, e.g. document
  frequency, font size, position
• Geoconfidence

• Aim combination reflects users preferred
  balance between recall and correctness of the
  geographic reference
• e.g. Georelevance query term relevance
  geoconfidence

18
Conclusion

• Ambiguity problem much worse with large
  gazetteers
• Can use probabilistic methods where feasible
  (local information), combine with
  confidence-based heuristics