CS224N Final Project Geo-location Route Recognition
description
Transcript of CS224N Final Project Geo-location Route Recognition
![Page 1: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/1.jpg)
CS224N Final ProjectGeo-location Route Recognition
Yingjie (Roger) ZhengPhilip (Tony) Hairr
June 9, 2010
![Page 2: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/2.jpg)
Objective
• We would like that our system can extract a list of locations from web pages that represents the direction of the route and plot the route on a map.
![Page 3: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/3.jpg)
Example
From www.lonelyplanet.com
![Page 4: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/4.jpg)
Pipeline
Crawler
Crawler
NERNER
ParserParser
Route Disambiguate Engine
Route Disambiguate Engine
Map RendererMap Renderer
Acquire webpage
Recognize place names & organization names
Get word dependencies
Arrange route
Get coordinates & draw map
![Page 5: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/5.jpg)
From Typed Dependency to RoutePrepositional Phrase
• I took a bus ride to Sacramento from Chicago.nsubj(took-2, I-1)
det(ride-5, a-3)nn(ride-5, bus-4)
dobj(took-2, ride-5)prep(took-2, to-6)
pobj(to-6, Sacramento-7)prep(took-2, from-8)
pobj(from-8, Chicago-9)
From To
Chicago Sacramento
![Page 6: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/6.jpg)
From Typed Dependency to RouteTransitive Verbs
• I left Palo Alto for New York this morning.
nsubj(left-2, I-1)dobj(left-2, Palo_Alto-3)
prep(Palo_Alto-3, for-4)pobj(for-4, New_York-5)
det(morning-7, this-6)tmod(left-2, morning-7)
From To
Palo Alto New York
![Page 7: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/7.jpg)
Evaluation
Score =
Precision: We generate lists of unique places appearing in the test program output and the golden test data separately, then match them to find out how many locations appear in both, then calculated precision using the matching and total line counts.
Recall: We calculate recall by dividing the matching lines by the total lines in the golden test data.
locations in the golden test data
locations in the golden test data + edit distance
![Page 8: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/8.jpg)
Test and Results
• Data• Forum data from www.lonelyplanet.com
• Baseline• Start and end point according to the order of
appearance
• Method• Look five sentences in a forum page
• Result
Precision Recall Score
Our system 0.549 0.602 0.438
Baseline 0.537 0.454 0.588
![Page 9: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/9.jpg)
Example Output
![Page 10: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/10.jpg)
Example Output Locations
Output Route
San Cristobal de las CasasSan Miguel de AllendeSan Cristobal de las CasasSan Miguel de AllendeOaxacaSan CristobalSan CristobalMexico CitySan Miguel
Golden Route
San Cristobal de las CasasTuxla GutierrezMexico CitySan Miguel de Allende
![Page 11: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/11.jpg)
Problems and Future Work
Precision and Recall of the NER system
How to recognize different routes in one document according to context
Location ambiguityCambridge: Cambridge, MA or Cambridge, UK
Crawler
Crawler
NERNER
ParserParser
Route Disambiguate Engine
Route Disambiguate Engine
Map RendererMap Renderer
![Page 12: CS224N Final Project Geo-location Route Recognition](https://reader036.fdocuments.us/reader036/viewer/2022070409/568144c3550346895db18d26/html5/thumbnails/12.jpg)
Thank you