Grounding Text Jason Baldridge @jasonbaldridge Austin Data 2014 Associate Professor Co-founder & Chief Scientist Friday, September 5, 14
Nov 29, 2014
Grounding Text
Jason Baldridge @jasonbaldridge
Austin Data 2014
Associate Professor Co-founder & Chief Scientist
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean?
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
What does “barbecue” mean? Barbecue’
2
Friday, September 5, 14
© 2012 Jason M Baldridge Text Analytics Summit, June 2013
What I thought semantics was before 2005
3
From: John Enrico and Jason Baldridge. 2011. Possessor Raising, Demonstrative Raising, Quantifier Float and Number Float in Haida. International Journal of American Linguistics. 77(2):185-218
Friday, September 5, 14
© 2012 Jason M Baldridge Text Analytics Summit, June 2013
Updated perspective a la Ray Mooney (UT Austin CS)
4
http://www.cs.utexas.edu/users/ml/slides/chen-icml08.ppt
Friday, September 5, 14
© 2012 Jason M Baldridge Text Analytics Summit, June 2013
http://www.lib.utexas.edu/books/travel/index.htmlTravel at the Turn of the 20th Century
5
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Motivation: Google Lit Trips [http://www.googlelittrips.com/]
6
Grapes of Wrath in Google Earth
Text
http://www.googlelittrips.com/GoogleLit/9-12/Entries/2006/11/1_The_Grapes_of_Wrath_by_John_Steinbeck.html
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Motivation: Google Lit Trips [http://www.googlelittrips.com/]
6
Grapes of Wrath in Google Earth
Text
http://www.googlelittrips.com/GoogleLit/9-12/Entries/2006/11/1_The_Grapes_of_Wrath_by_John_Steinbeck.html
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Crisis response: Haiti earthquake
7
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Crisis response: Haiti earthquake
7
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Look, Mom, no hands! (Err, um... no metadata.)
8
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Look, Mom, no hands! (Err, um... no metadata.)
8
Topics with a clear, circumscribed geographic focus emerge!
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
But, of course, metadata is now plentiful.
9
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Geotagged Wikipedia
10
30° 17′ N 97° 44′ W
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
01:55:55 RT @USER_dc5e5498: Drop and give me 50....
05:09:29 I said u got a swisher from redmond!? He said nah kirkland! Lol..ooooooooOkay!
05:57:35 Lmao!:) havin a good ol time after work! Unexpected! #goodtimes
06:00:09 RT @USER_d5d93fec: #letsbereal .. No seriously, #letsbereal>>lol. Don't start.
06:00:37 On my way to get @USER_60939380 yeee! She want some of this strawberry! Sexy!
...
47°31’41’’ N 122°11’52’’ W11
Geotagged Twitter
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
01:55:55 RT @USER_dc5e5498: Drop and give me 50....
05:09:29 I said u got a swisher from redmond!? He said nah kirkland! Lol..ooooooooOkay!
05:57:35 Lmao!:) havin a good ol time after work! Unexpected! #goodtimes
06:00:09 RT @USER_d5d93fec: #letsbereal .. No seriously, #letsbereal>>lol. Don't start.
06:00:37 On my way to get @USER_60939380 yeee! She want some of this strawberry! Sexy!
...
47°31’41’’ N 122°11’52’’ W11
Geotagged Twitter
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Document geolocation: where is this person?
12
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 201313
Amsterdam, Zaandam, Amstelveen, Diemen, Landsmeer ...
Frankfurt, Frechen, Hürth, Brühl, Wesseling, ...
Language modeling approach
Wing & Baldridge 2011: Simple supervised document geolocation with geodesic grids.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 201313
Amsterdam, Zaandam, Amstelveen, Diemen, Landsmeer ...
Frankfurt, Frechen, Hürth, Brühl, Wesseling, ...
Language modeling approach
Wing & Baldridge 2011: Simple supervised document geolocation with geodesic grids.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
mountainbeach
wine barbecue
Where’s a word on Earth?
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
mountainbeach
wine barbecue
Where’s a word on Earth?
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Locations of Twitter users are not uniformly distributed!
15
(Small) GeoUT (Twitter) plotted on Google Earth, one pin per user.
Density of (all) documents in GeoUT
over the USA (390 million tweets)
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
k-d tree for geotagged Wikipedia, looking at N. America
16
Roller, Speriosu, Rallapalli, Wing & Baldridge 2014: Supervised Text-based Geolocation Using Language Models on an Adaptive Grid.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
k-d tree for geotagged Wikipedia, looking at N. America
16
Roller, Speriosu, Rallapalli, Wing & Baldridge 2014: Supervised Text-based Geolocation Using Language Models on an Adaptive Grid.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Pre-grid clustering [Erik Skiles, MA thesis, UT Austin, Ling]
17
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Four clusters on GeoUT (390 million tweets)
18
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Four clusters on GeoUT (390 million tweets)
18
West coast East coast Midwest & South Spanish language
All tweets
Friday, September 5, 14
[Serdyukov, Murdock, & van Zwol 2009; Cheng, Caverlee, & Lee 2010; Wing & Baldridge 2011]
Automatic document geolocation
Friday, September 5, 14
[Serdyukov, Murdock, & van Zwol 2009; Cheng, Caverlee, & Lee 2010; Wing & Baldridge 2011]
Automatic document geolocation
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Image geo-location: http://graphics.cs.cmu.edu/projects/im2gps/
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Performance (kd-tree with clustering)
21
Wikipedia (entire world)Half of documents geotagged within 12 km of truthPercent of documents within 166km (100 miles): 91%
Twitter (USA)Half of users geotagged within 330 km of truthPercent of documents within 166km (100 miles): 40%
For better or worse, it soon might not matter whether you have location turned on or not... what
you say is where you are / are from. (Also, other factors, e.g. who you are linked to, of course.)
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Hierarchical geo-location with logistic regression
22
Wing & Baldridge 2014: Hierarchical Discriminative Classification for Text-Based Geolocation.
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Performance (kd-tree with clustering)
23
Flickr (entire world)Half of documents geotagged within 18 km of truthPercent of documents within 166km (100 miles): 66%
Twitter (World)Half of users geotagged within 490 km of truthPercent of documents within 166km (100 miles): 31%
Twitter (USA)Half of users geotagged within 170 km of truthPercent of documents within 166km (100 miles): 49%
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Hierarchical logistic regression beats flat naive Bayes
24
Naive Bayes Hierarchical LR
Twitter USA
Twitter World
Flickr
English Wikipedia
German Wikipedia
Portuguese Wikipedia
36.2 49.2
28.7 31.3
58.5 66.0
84.5 88.9
89.3 90.2
77.1 89.5
Accuracy @ 161 km, kd-tree grid
Friday, September 5, 14
© 2010 Jason M Baldridge Text Analytics Summit, June 2013
Logistic regression weights good features heavily
25
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym (place name) resolution
26
They visit Portland every year.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym (place name) resolution
26
They visit Portland every year.
?
?
?
?
?
?
?
?
?
?
?
?
?
?
??
?
Which Portland? (Also: Canada, Australia, Ireland...)
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym resolution in context
27
Although Elisha Newman made the first land entry in the township of Portland (June, 1833), he did not become a settler until three years later, by which time a few settlers had located in the town. From Mr. Newman's story, it appears that early in 1833, he was visiting friends in Ann Arbor, and during an evening conversation discussed with others the subject of unlocated lands lying west of Ann Arbor. One of the company (Joseph Wood) remarked that he had been out with the party sent to survey Ionia and other counties, and that the surveyors were struck by the valuable water-power at the mouth of the Looking Glass River, saying there would surely be a village there some day.Mr. Newman was at once taken with the idea of locating lands at the mouth of the Looking Glass. Following up his impulse, he made ready to start at once, and, accompanied by James Newman and Joseph Wood, went out to the Looking Glass on a tour of inspection. Being satisfied with the location, he returned Eastward with his companions, and at White Pigeon made his land entry.Newman did not return for a permanent settlement until the spring of 1836, and meanwhile, in November, 1833, Philo Bogue bought a piece of land on section 28, in the bend of the Grand River, where he proposed to set up a trading post. Unaided he rolled up a log cabin near where the Detroit, Lansing, and Northern depot was located, and when he brought the house into decent shape went over to Hunt's at Lyons for his family, whom he had left there against such time as he should have affairs prepared for their comfort.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Spatial minimality
28
Although Elisha Newman made the first land entry in the township of Portland (June, 1833), he did not become a settler until three years later, by which time a few settlers had located in the town. From Mr. Newman's story, it appears that early in 1833, he was visiting friends in Ann Arbor, and during an evening conversation discussed with others the subject of unlocated lands lying west of Ann Arbor. One of the company (Joseph Wood) remarked that he had been out with the party sent to survey Ionia and other counties, and that the surveyors were struck by the valuable water-power at the mouth of the Looking Glass River, saying there would surely be a village there some day.
Mr. Newman was at once taken with the idea of locating lands at the mouth of the Looking Glass. Following up his impulse, he made ready to start at once, and, accompanied by James Newman and Joseph Wood, went out to the Looking Glass on a tour of inspection. Being satisfied with the location, he returned Eastward with his companions, and at White Pigeon made his land entry.
Newman did not return for a permanent settlement until the spring of 1836, and meanwhile, in November, 1833, Philo Bogue bought a piece of land on section 28, in the bend of the Grand River, where he proposed to set up a trading post. Unaided he rolled up a log cabin near where the Detroit, Lansing, and Northern depot was located, and when he brought the house into decent shape went over to Hunt's at Lyons for his family, whom he had left there against such time as he should have affairs prepared for their comfort.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Geo
Nam
es
4048392 Portland Mills Portland Mills 39.7781 -87.00918 P PPL US IN 133 0 223 218 America/Indiana/Indianapolis 2010-02-154084605 Portland Portland 32.15459 -87.1686 P PPL US AL 047 0 30 41 America/Chicago 2006-01-154127143 Portland Portland Portlend,Портленд 33.2379 -91.51151 P PPL US AR 003 430 38 39 America/Chicago 2011-05-144169227 Portland Portland 30.51242 -86.19578 P PPL US FL 131 0 8 14 America/Chicago 2006-01-154217115 Portland Portland 34.05732 -85.03634 P PPL US GA 233 0 229 228 America/New_York 2010-09-054277586 Portland Portland 37.0778 -97.31227 P PPL US KS 191 0 362 364 America/Chicago 2006-01-154305000 Portland Portland 37.12062 -85.44608 P PPL US KY 001 0 220 223 America/Chicago 2006-01-154305001 Portland Portland 38.26924 -85.8108 P PPL US KY 111 0 135 138 America/Kentucky/Louisville 2006-01-154305002 Portland Portland 38.74812 -84.44772 P PPL US KY 191 0 265 266 America/New_York 2006-01-15404289 Portland Portland Portlend,Портленд 38.71088 -91.71767 P PPL US MO 027 0 170 172 America/Chicago 2010-01-294521811 Portland Portland Portlend,Портленд 39.00341 -81.77124 P PPL US OH 105 0 187 188 America/New_York 2010-01-294650946 Portland Portland Portlend,Портленд 36.58171 -86.51638 P PPL US TN 165 11480 244 245 America/Chicago 2011-05-144720131 Portland Portland Portlend,Портленд 27.87725 -97.32388 P PPL US TX 409 15099 13 11 America/Chicago 2011-05-144841001 Portland Portland Portlend,Портленд 41.57288 -72.64065 P PPL US CT 007 5862 24 27 America/New_York 2011-05-144871855 Portland Portland 43.12858 -93.12354 P PPL US IA 033 35 327 330 America/Chicago 2011-05-144906524 Portland Portland 41.66253 -89.98012 P PPL US IL 195 0 190 190 America/Chicago 2006-01-155006314 Portland Portland Portlend,Портленд 42.8692 -84.90305 P PPL US MI 067 3883 221 223 America/Detroit 2011-05-145746545 Portland Portland 45.52345 -122.67621 P PPLA2 US OR 051 583776 12 15 America/Los_Angeles 2011-05-14
Spatial minimality
28
Although Elisha Newman made the first land entry in the township of Portland (June, 1833), he did not become a settler until three years later, by which time a few settlers had located in the town. From Mr. Newman's story, it appears that early in 1833, he was visiting friends in Ann Arbor, and during an evening conversation discussed with others the subject of unlocated lands lying west of Ann Arbor. One of the company (Joseph Wood) remarked that he had been out with the party sent to survey Ionia and other counties, and that the surveyors were struck by the valuable water-power at the mouth of the Looking Glass River, saying there would surely be a village there some day.
Mr. Newman was at once taken with the idea of locating lands at the mouth of the Looking Glass. Following up his impulse, he made ready to start at once, and, accompanied by James Newman and Joseph Wood, went out to the Looking Glass on a tour of inspection. Being satisfied with the location, he returned Eastward with his companions, and at White Pigeon made his land entry.
Newman did not return for a permanent settlement until the spring of 1836, and meanwhile, in November, 1833, Philo Bogue bought a piece of land on section 28, in the bend of the Grand River, where he proposed to set up a trading post. Unaided he rolled up a log cabin near where the Detroit, Lansing, and Northern depot was located, and when he brought the house into decent shape went over to Hunt's at Lyons for his family, whom he had left there against such time as he should have affairs prepared for their comfort.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Geo
Nam
es
4048392 Portland Mills Portland Mills 39.7781 -87.00918 P PPL US IN 133 0 223 218 America/Indiana/Indianapolis 2010-02-154084605 Portland Portland 32.15459 -87.1686 P PPL US AL 047 0 30 41 America/Chicago 2006-01-154127143 Portland Portland Portlend,Портленд 33.2379 -91.51151 P PPL US AR 003 430 38 39 America/Chicago 2011-05-144169227 Portland Portland 30.51242 -86.19578 P PPL US FL 131 0 8 14 America/Chicago 2006-01-154217115 Portland Portland 34.05732 -85.03634 P PPL US GA 233 0 229 228 America/New_York 2010-09-054277586 Portland Portland 37.0778 -97.31227 P PPL US KS 191 0 362 364 America/Chicago 2006-01-154305000 Portland Portland 37.12062 -85.44608 P PPL US KY 001 0 220 223 America/Chicago 2006-01-154305001 Portland Portland 38.26924 -85.8108 P PPL US KY 111 0 135 138 America/Kentucky/Louisville 2006-01-154305002 Portland Portland 38.74812 -84.44772 P PPL US KY 191 0 265 266 America/New_York 2006-01-15404289 Portland Portland Portlend,Портленд 38.71088 -91.71767 P PPL US MO 027 0 170 172 America/Chicago 2010-01-294521811 Portland Portland Portlend,Портленд 39.00341 -81.77124 P PPL US OH 105 0 187 188 America/New_York 2010-01-294650946 Portland Portland Portlend,Портленд 36.58171 -86.51638 P PPL US TN 165 11480 244 245 America/Chicago 2011-05-144720131 Portland Portland Portlend,Портленд 27.87725 -97.32388 P PPL US TX 409 15099 13 11 America/Chicago 2011-05-144841001 Portland Portland Portlend,Портленд 41.57288 -72.64065 P PPL US CT 007 5862 24 27 America/New_York 2011-05-144871855 Portland Portland 43.12858 -93.12354 P PPL US IA 033 35 327 330 America/Chicago 2011-05-144906524 Portland Portland 41.66253 -89.98012 P PPL US IL 195 0 190 190 America/Chicago 2006-01-155006314 Portland Portland Portlend,Портленд 42.8692 -84.90305 P PPL US MI 067 3883 221 223 America/Detroit 2011-05-145746545 Portland Portland 45.52345 -122.67621 P PPLA2 US OR 051 583776 12 15 America/Los_Angeles 2011-05-14
Spatial minimality
28
Ann ArborDetroit
IoniaLyons
PortlandWhite Pigeon
1>7>4
>15>17
1
# LocationsToponym
Although Elisha Newman made the first land entry in the township of Portland (June, 1833), he did not become a settler until three years later, by which time a few settlers had located in the town. From Mr. Newman's story, it appears that early in 1833, he was visiting friends in Ann Arbor, and during an evening conversation discussed with others the subject of unlocated lands lying west of Ann Arbor. One of the company (Joseph Wood) remarked that he had been out with the party sent to survey Ionia and other counties, and that the surveyors were struck by the valuable water-power at the mouth of the Looking Glass River, saying there would surely be a village there some day.
Mr. Newman was at once taken with the idea of locating lands at the mouth of the Looking Glass. Following up his impulse, he made ready to start at once, and, accompanied by James Newman and Joseph Wood, went out to the Looking Glass on a tour of inspection. Being satisfied with the location, he returned Eastward with his companions, and at White Pigeon made his land entry.
Newman did not return for a permanent settlement until the spring of 1836, and meanwhile, in November, 1833, Philo Bogue bought a piece of land on section 28, in the bend of the Grand River, where he proposed to set up a trading post. Unaided he rolled up a log cabin near where the Detroit, Lansing, and Northern depot was located, and when he brought the house into decent shape went over to Hunt's at Lyons for his family, whom he had left there against such time as he should have affairs prepared for their comfort.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Geo
Nam
es
4048392 Portland Mills Portland Mills 39.7781 -87.00918 P PPL US IN 133 0 223 218 America/Indiana/Indianapolis 2010-02-154084605 Portland Portland 32.15459 -87.1686 P PPL US AL 047 0 30 41 America/Chicago 2006-01-154127143 Portland Portland Portlend,Портленд 33.2379 -91.51151 P PPL US AR 003 430 38 39 America/Chicago 2011-05-144169227 Portland Portland 30.51242 -86.19578 P PPL US FL 131 0 8 14 America/Chicago 2006-01-154217115 Portland Portland 34.05732 -85.03634 P PPL US GA 233 0 229 228 America/New_York 2010-09-054277586 Portland Portland 37.0778 -97.31227 P PPL US KS 191 0 362 364 America/Chicago 2006-01-154305000 Portland Portland 37.12062 -85.44608 P PPL US KY 001 0 220 223 America/Chicago 2006-01-154305001 Portland Portland 38.26924 -85.8108 P PPL US KY 111 0 135 138 America/Kentucky/Louisville 2006-01-154305002 Portland Portland 38.74812 -84.44772 P PPL US KY 191 0 265 266 America/New_York 2006-01-15404289 Portland Portland Portlend,Портленд 38.71088 -91.71767 P PPL US MO 027 0 170 172 America/Chicago 2010-01-294521811 Portland Portland Portlend,Портленд 39.00341 -81.77124 P PPL US OH 105 0 187 188 America/New_York 2010-01-294650946 Portland Portland Portlend,Портленд 36.58171 -86.51638 P PPL US TN 165 11480 244 245 America/Chicago 2011-05-144720131 Portland Portland Portlend,Портленд 27.87725 -97.32388 P PPL US TX 409 15099 13 11 America/Chicago 2011-05-144841001 Portland Portland Portlend,Портленд 41.57288 -72.64065 P PPL US CT 007 5862 24 27 America/New_York 2011-05-144871855 Portland Portland 43.12858 -93.12354 P PPL US IA 033 35 327 330 America/Chicago 2011-05-144906524 Portland Portland 41.66253 -89.98012 P PPL US IL 195 0 190 190 America/Chicago 2006-01-155006314 Portland Portland Portlend,Портленд 42.8692 -84.90305 P PPL US MI 067 3883 221 223 America/Detroit 2011-05-145746545 Portland Portland 45.52345 -122.67621 P PPLA2 US OR 051 583776 12 15 America/Los_Angeles 2011-05-14
Spatial minimality
28
PortlandLyonsIonia
White Pigeon
Ann ArborDetroit
IoniaLyons
PortlandWhite Pigeon
1>7>4
>15>17
1
# LocationsToponym
Although Elisha Newman made the first land entry in the township of Portland (June, 1833), he did not become a settler until three years later, by which time a few settlers had located in the town. From Mr. Newman's story, it appears that early in 1833, he was visiting friends in Ann Arbor, and during an evening conversation discussed with others the subject of unlocated lands lying west of Ann Arbor. One of the company (Joseph Wood) remarked that he had been out with the party sent to survey Ionia and other counties, and that the surveyors were struck by the valuable water-power at the mouth of the Looking Glass River, saying there would surely be a village there some day.
Mr. Newman was at once taken with the idea of locating lands at the mouth of the Looking Glass. Following up his impulse, he made ready to start at once, and, accompanied by James Newman and Joseph Wood, went out to the Looking Glass on a tour of inspection. Being satisfied with the location, he returned Eastward with his companions, and at White Pigeon made his land entry.
Newman did not return for a permanent settlement until the spring of 1836, and meanwhile, in November, 1833, Philo Bogue bought a piece of land on section 28, in the bend of the Grand River, where he proposed to set up a trading post. Unaided he rolled up a log cabin near where the Detroit, Lansing, and Northern depot was located, and when he brought the house into decent shape went over to Hunt's at Lyons for his family, whom he had left there against such time as he should have affairs prepared for their comfort.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Spatial minimality often fails
29
I moved from Encinitas, CA, a nice beach town in North San Diego County to Asheville, NC. By far, Ashville is more hip, especially West Asheville. Asheville has a lot in common with Portland. Austin, I've never been to so I cannot comment. But what makes a place cool and hip, in my opinion are that give a area "punch". There are a lot of ingredients. One is geography. Add a college or university (and all that they bring- and draw), good restaurants, a good music scene, a progressive attitude and tolerance. Hmmm. I'm sure there are many more to ponder. But that's my start. Oh, lots of bars!
From: http://www.city-data.com/forum/austin/1694181-what-makes-city-like-austin-portland-3.html
City-data.com incorrectly marks “West” and “Portland” as the cities in Texas -- presumably because of their textual and spatial proximity to “Austin”.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Spatial minimality often fails
29
I moved from Encinitas, CA, a nice beach town in North San Diego County to Asheville, NC. By far, Ashville is more hip, especially West Asheville. Asheville has a lot in common with Portland. Austin, I've never been to so I cannot comment. But what makes a place cool and hip, in my opinion are that give a area "punch". There are a lot of ingredients. One is geography. Add a college or university (and all that they bring- and draw), good restaurants, a good music scene, a progressive attitude and tolerance. Hmmm. I'm sure there are many more to ponder. But that's my start. Oh, lots of bars!
From: http://www.city-data.com/forum/austin/1694181-what-makes-city-like-austin-portland-3.html
City-data.com incorrectly marks “West” and “Portland” as the cities in Texas -- presumably because of their textual and spatial proximity to “Austin”.
But: it is clear from the text that Portland, Oregon and Austin, Texas are the referents, though their states are never mentioned and are far from the other locations!
I moved from Encinitas, CA, a nice beach town in North San Diego County to Asheville, NC. By far, Ashville is more hip, especially West Asheville. Asheville has a lot in common with Portland. Austin, I've never been to so I cannot comment. But what makes a place cool and hip, in my opinion are that give a area "punch". There are a lot of ingredients. One is geography. Add a college or university (and all that they bring- and draw), good restaurants, a good music scene, a progressive attitude and tolerance. Hmmm. I'm sure there are many more to ponder. But that's my start. Oh, lots of bars!
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Toponym classifiers
30
Strategy: build a textual classifier per toponym by obtaining indirectly labeled examples from Wikipedia.
P(Portland-OR|music) > P(Portland-ME|music)P(Portland-OR|wharf ) < P(Portland-ME|wharf )
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Results: disambiguating toponyms
31
Average error distance
Accuracy Average error distance
Accuracy
Population
SPIDER(spatial minimality)
WISTR(Wiki supervised)
SPIDER+WISTR
216 81.0 1749 59.7
2180 30.9 266 57.5
279 82.3 855 69.1
430 81.8 201 85.9
TR-CoNLLReuters News Texts
August 1996
Perseus Civil War CorpusBooks
Late 19th Century
Take-home message: text classifiers are very effective & can be boosted by spatial minimality algorithms.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Identifying, disambiguating, and displaying toponyms
32
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Back to grounding
33
Grounding often involves connecting text to knowledge sources and other modalities (image, video) & bootstrapping.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Back to grounding
33
Grounding often involves connecting text to knowledge sources and other modalities (image, video) & bootstrapping.
Also, they can help us create models for deeper aspects of language, such as syntactic structure and logical form.
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Lexical brain decoding [Yarkoni, Poldrack, Nichols, Van Essen & Wager (2011)]
34
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Lexical brain decoding [Yarkoni, Poldrack, Nichols, Van Essen & Wager (2011)]
34
Friday, September 5, 14
He says, she says http://www.tweetolife.com/gender/
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporality of words, by hour http://www.tweetolife.com/hour/
36
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporality of words, by hour http://www.tweetolife.com/hour/
36
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporality of expressions, by day: http://www.google.com/trends
37
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporality of expressions, by day: http://www.google.com/trends
37
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporality of expressions, by year: http://ngrams.googlelabs.com/
38
slavetrenches aircraft
war
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Temporal resolution [Kumar, Lease, and Baldridge 2011]
39
2000
BC
0 A
D
2000
AD
4000
BC
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
More modalities: videos [Motwani & Mooney, 2012]
40
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Beyond word co-occurences for vector-space models
41
bear boat car cow hadoop snow water wrench
3 234 42 4 1 2 325 0
beach
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Combining distributional models with logics
42
Erk (2013): “Towards a semantics for distributional representations.”
Garrette et al (2012): “A formal approach to linking logical form and vector-space lexical semantics”Beltagy et al (2013): “Montague Meets Markov: Deep Semantics with Probabilistic Logical Form”
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Multi-component structured vector-space models
43
beachchildren
visit
the children visit the beach
Agent Patient
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Language learning in context [Kim & Mooney, 2013]
44
Friday, September 5, 14
© 2013 Jason M Baldridge Text Analytics Summit, June 2013
Language learning in context [Kim & Mooney, 2013]
44
Friday, September 5, 14
All your meaning are belong to us
Friday, September 5, 14
All your meaning are belong to us
Friday, September 5, 14
All your meaning are belong to us
Friday, September 5, 14
http://davidrothman.net/2009/09/02/all-your-healthbase-are-belong-to-us-want-em-back/
Grounding matters
Friday, September 5, 14
Junto - label propagationhttps://github.com/scalanlp/junto
Textgrounder - document geolocationhttps://github.com/utcompling/textgrounder
Fieldspring - toponym resolutionhttps://github.com/utcompling/fieldspring
Low-resource POS tagginghttps://github.com/dhgarrette/low-resource-
pos-tagging-2013
Updown - polarity classificationhttps://github.com/scalanlp/junto
OpenNLP - machine learning / NLPhttp://opennlp.apache.org/
Open Source Software (Scala/Java)
Friday, September 5, 14
Junto - label propagationhttps://github.com/scalanlp/junto
Textgrounder - document geolocationhttps://github.com/utcompling/textgrounder
Fieldspring - toponym resolutionhttps://github.com/utcompling/fieldspring
Low-resource POS tagginghttps://github.com/dhgarrette/low-resource-
pos-tagging-2013
Updown - polarity classificationhttps://github.com/scalanlp/junto
OpenNLP - machine learning / NLPhttp://opennlp.apache.org/
Nak - machine learninghttps://github.com/scalanlp/nak
Chalk - NLPhttps://github.com/scalanlp/chalk
Breeze - linear algebrahttps://github.com/scalanlp/nak
Scal
aNLP
Open Source Software (Scala/Java)
Friday, September 5, 14
Junto - label propagationhttps://github.com/scalanlp/junto
Textgrounder - document geolocationhttps://github.com/utcompling/textgrounder
Fieldspring - toponym resolutionhttps://github.com/utcompling/fieldspring
Low-resource POS tagginghttps://github.com/dhgarrette/low-resource-
pos-tagging-2013
Updown - polarity classificationhttps://github.com/scalanlp/junto
OpenNLP - machine learning / NLPhttp://opennlp.apache.org/
Nak - machine learninghttps://github.com/scalanlp/nak
Chalk - NLPhttps://github.com/scalanlp/chalk
Breeze - linear algebrahttps://github.com/scalanlp/nak
Scal
aNLP
Open Source Software (Scala/Java)
Friday, September 5, 14
This research was sponsored by:
Grant: W911NF-10-1-0533Grant from the Morris Memorial Trust Fund
- Walt Whitman, A Song of the Rolling Earth (in Leaves of Grass)
Final note: Whitman had it right many years ago!
Friday, September 5, 14
Supervision- documents labeled with latitude & longitude
Methods- Language Modeling for Information Retrieval
Code- Textgrounder: https://github.com/utcompling/textgrounder
Publications- Stephen Roller, Mike Speriosu, Sarat Rallapalli, Benjamin Wing and Jason Baldridge. 2012. Supervised Text-based Geolocation Using Language Models on an Adaptive Grid. EMNLP 2012. Jeju, Korea.- Benjamin Wing and Jason Baldridge. 2011. Simple supervised document geolocation with geodesic grids. In Proceedings of ACL HLT 2011.
Document geolocation
Friday, September 5, 14
Supervision- indirectly acquired toponym annotations using a gazeteer and geo-annotated Wikipedia
Methods- logistic regression- named entity recognition
Code- Fieldspring: https://github.com/utcompling/fieldspring
Publications- Mike Speriosu and Jason Baldridge. Text-Driven Toponym Resolution using Indirect Supervision. To appear in proceedings of ACL 2013.
Toponym resolution
Friday, September 5, 14