Using Machine Learning to Discover and Understand Structured Data William W. Cohen Machine Learning Dept. and Language Technologies Inst. School of Computer

Using Machine Learning to Discover and Understand Structured Data

William W. Cohen

Machine Learning Dept. and Language Technologies Inst.School of Computer ScienceCarnegie Mellon University

Outline

• Information integration: – Some history– The problem, the economics, and the

economic problem

• “Soft” information integration• Concrete uses of “soft” integration

– Classification– Collaborative filtering– Set expansion

When are two entities the same?

• Bell Labs• Bell Telephone Labs• AT&T Bell Labs• A&T Labs• AT&T Labs—Research• AT&T Labs Research,

Shannon Laboratory• Shannon Labs• Bell Labs Innovations• Lucent Technologies/Bell

Labs Innovations

History of Innovation: From 1925 to today, AT&T has attracted some of the world's greatest scientists, engineers and developers…. [www.research.att.com]

Bell Labs Facts: Bell Laboratories, the research and development arm of Lucent Technologies, has been operating continuously since 1925… [bell-labs.com]

[1925]

In the once upon a time days of the First Age of Magic, the prudent sorcerer regarded his own true name as his most valued possession but also the greatest threat to his continued good health, for--the stories go--once an enemy, even a weak unskilled enemy, learned the sorcerer's true name, then routine and widely known spells could destroy or enslave even the most powerful. As times passed, and we graduated to the Age of Reason and thence to the first and second industrial revolutions, such notions were discredited. Now it seems that the Wheel has turned full circle (even if there never really was a First Age) and we are back to worrying about true names again:

The first hint Mr. Slippery had that his own True Name might be known--and, for that matter, known to the Great Enemy--came with the appearance of two black Lincolns humming up the long dirt driveway ... Roger Pollack was in his garden weeding, had been there nearly the whole morning.... Four heavy-set men and a hard-looking female piled out, started purposefully across his well-tended cabbage patch.…

This had been, of course, Roger Pollack's great fear. They had discovered Mr. Slippery's True Name and it was Roger Andrew Pollack TIN/SSAN 0959-34-2861.

Deduction via co-operation

Site1Site2

Site3

KB1KB2

KB3

Standard Terminology

Integrated KB

UserEconomic issues:

• Who pays for integration? Who tracks errors & inconsistencies? Who fixes bugs? Who pushes for clarity in underlying concepts and object identifiers?

• Standards approach publishers are responsible publishers pay

• Mediator approach: 3rd party does the work, agnostic as to cost

LinkageQueries

Traditional approach:

Uncertainty about what to linkmust be decided by the integration

system, not the end user

Link items asneeded by Q

Query Q

SELECT R.a,S.a,S.b,T.b FROM R,S,T

WHERE R.a=S.a and S.b=T.b

R.a S.a S.b T.b

Anhai Anhai Doan Doan

Dan Dan Weld Weld

Strongest links: those agreeable to most users

William Will Cohen Cohn

Steve Steven Minton Mitton

Weaker links: those agreeable to some users

William David Cohen Cohneven weaker links…

WHIRL approach:

Link items asneeded by Q

WHIRL approach:

Query Q

SELECT R.a,S.a,S.b,T.b FROM R,S,T

WHERE R.a~S.a and S.b~T.b (~ TFIDF-similar)

R.a S.a S.b T.b

Anhai Anhai Doan Doan

Dan Dan Weld Weld

Incrementally produce a ranked list of possible links,

with “best matches” first. User (or downstream process)

decides how much of the list to generate and examine.

William Will Cohen Cohn

Steve Steven Minton Mitton

William David Cohen Cohn

WHIRL queries• Assume two relations:

review(movieTitle,reviewText): archive of reviews

listing(theatre, movieTitle, showTimes, …): now showing

The Hitchhiker’s Guide to the Galaxy, 2005

This is a faithful re-creation of the original radio series – not surprisingly, as Adams wrote the screenplay ….

Men in Black, 1997

Will Smith does an excellent job in this …

Space Balls, 1987

Only a die-hard Mel Brooks fan could claim to enjoy …

… …

Star Wars Episode III

The Senator Theater

1:00, 4:15, & 7:30pm.

Cinderella Man

The Rotunda Cinema

1:00, 4:30, & 7:30pm.

… … …

WHIRL queries

• “Find reviews of sci-fi comedies [movie domain]

FROM review SELECT * WHERE r.text~’sci fi comedy’

(like standard ranked retrieval of “sci-fi comedy”)

• “ “Where is [that sci-fi comedy] playing?”FROM review as r, LISTING as s, SELECT *

WHERE r.title~s.title and r.text~’sci fi comedy’

(best answers: titles are similar to each other – e.g., “Hitchhiker’s Guide to the Galaxy” and “The Hitchhiker’s Guide to the Galaxy, 2005” and the review text is similar to “sci-fi comedy”)

WHIRL queries• Similarity is based on TFIDF rare words are most important.

• Search for high-ranking answers uses inverted indices….

The Hitchhiker’s Guide to the Galaxy, 2005

Men in Black, 1997

Space Balls, 1987

…

Star Wars Episode III

Hitchhiker’s Guide to the Galaxy

Cinderella Man

…

Years are common in the review archive, so have low weight

hitchhiker movie00137

the movie001,movie003,movie007,movie008, movie013,movie018,movie023,movie0031,

…..

- It is easy to find the (few) items that match on “important” terms

- Search for strong matches can prune “unimportant terms”

Outline


economic problem



Outline


economic problem



Outline

• Information integration: – Some history– The problem, the economics, and the economic

problem


– Classification– Collaborative filtering– Set expansion: using generalized notion of

similarity

Recent work: non-textual similarity

“William W. Cohen, CMU”

“Dr. W. W. Cohen”

cohenwilliam w

drcmu

“George W. Bush”

“George H. W. Bush”

“Christos Faloutsos, CMU”

Recent work

• Personalized PageRank aka Random Walk with Restart:– Similarity measure for nodes in a graph, analogous

to TFIDF for text in a WHIRL database– natural extension to PageRank– amenable to learning parameters of the walk

(gradient search, w/ various optimization metrics):

• Toutanova, Manning & NG, ICML2004; Nie et al, WWW2005; Xi et al, SIGIR 2005

– various speedup techniques exist– queries:

Given type t* and node x, find y:T(y)=t* and y~x

proposal

CMU

CALO

graph

William

6/18/07

6/17/07

Sent To

Term In Subject

[email protected]

Learning to Search Email

[SIGIR 2006, CEAS 2006, WebKDD/SNA 2007]

Einat Minkov, CMU; Andrew Ng, Stanford

Tasks that are like similarity queries

Person namePerson namedisambiguationdisambiguation

ThreadingThreading

Alias findingAlias finding

[ term “andy” file msgId ]

“person”

[ file msgId ]

“file”

What are the adjacent messages in this thread?

A proxy for finding “more messages like this one”

What are the email-addresses of Jason ?...

[ term Jason ]

“email-address”

Meeting Meeting attendees finderattendees finder

Which email-addresses (persons) should I notify about this meeting? [ meeting mtgId ]

“email-address”

Learning to search better

Query a

node rank 1

node rank 2

node rank 3

node rank 4

…

node rank 10

node rank 11

node rank 12

…

node rank 50

Query b Query q

node rank 1

node rank 2

node rank 3

node rank 4

…

node rank 10

node rank 11

node rank 12

…

node rank 50

node rank 1

node rank 2

node rank 3

node rank 4

…

node rank 10

node rank 11

node rank 12

…

node rank 50

…

GRAPH WALK

+ Rel. answers a + Rel. answers b + Rel. answers q

Task T (query class)

Graph walk

Feature generatio

n

Learnre-ranker

Re-rankingfunction

Learning

Node re-ordering:

train task

Learning Approach

train task Graph walk

Feature generatio

n

Learnre-ranker

Re-rankingfunction

Graph walk

Feature generatio

n

Score byre-

ranking function

Node re-ordering:

Boosting

test task

[Collins & Koo, CL 2005; Collins, ACL 2002]

Voted Perceptron;RankSVM;

PerceptronCommittees;…

[Joacchim KDD 2002,Elsas et al WSDM 2008]

Graph walk

Weightupdate

Theta*

Learning approaches

Edge weight tuning:

Graph walk

Weightupdate

Graph walk

Learning approaches

Edge weight tuning:

Theta*

task

Graph walk

Feature generatio

n

Learnre-ranker

Re-rankingfunction

Graph walk

Feature generatio

n

Score byre-

ranking function

Node re-ordering:

Boosting;Voted Perceptron

Question: which is better?

[Diligenti et al, IJCAI 2005; Toutanova & Ng, ICML 2005; … ]

Results on one task

0%

20%

40%

60%

80%

100%

1 2 3 4 5 6 7 8 9 10

Rank

Rec

all

Mgmt. game

PER

SO

N

NA

ME

DIS

AM

BIG

UA

TIO

N

Results on several tasks (MAP)

0.4

0.45

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

M.game sager Shapiro

0.4

0.45

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

M.game Farmer Germany

0.4

0.45

0.5

0.55

0.6

0.65

0.7

0.75

0.8

0.85

Meetings

Namedisambiguation

Threading

Alias finding

*

*

*

*

*

*

*

*

** *

+

+

+

+ +

*

Set Expansion using the Web

• Fetcher: download web pages from the Web• Extractor: learn wrappers from web pages• Ranker: rank entities extracted by wrappers

1. Canon2. Nikon3. Olympus

4. Pentax5. Sony6. Kodak7. Minolta8. Panasonic9. Casio10. Leica11. Fuji12. Samsung13. …

Richard Wang, CMU

The Extractor

• Learn wrappers from web documents and seeds on the fly– Utilize semi-structured documents– Wrappers defined at character level

• No tokenization required; thus language independent

• However, very specific; thus page-dependent– Wrappers derived from document d is applied to d only

<img src="/common/logos/honda/logo-horiz-rgb-lg-dkbg.gif" alt="4"></a> <ul><li><a href="http://www.curryhonda-ga.com/"> Curry Honda Atlanta...</li> <li><a href="http://www.curryhondamass.com/"> Curry Honda...</li> <li class="last"><a href="http://www.curryhondany.com/"> Curry Honda Yorktown...</li></ul> </li>

<li class="honda"><a href="http://www.curryauto.com/">

<li class="acura"><a href="http://www.curryauto.com/">

<li class="toyota"><a href="http://www.curryauto.com/">

<li class="nissan"><a href="http://www.curryauto.com/">

<li class="ford"><a href="http://www.curryauto.com/"> <img src="/common/logos/ford/logo-horiz-rgb-lg-dkbg.gif" alt="3"></a> <ul><li class="last"><a href="http://www.curryauto.com/"> Curry Ford...</li></ul> </li>

<img src="/curryautogroup/images/logo-horiz-rgb-lg-dkbg.gif" alt="5"></a> <ul><li class="last"><a href="http://www.curryacura.com/"> Curry Acura...</li></ul> </li>

<img src="/common/logos/toyota/logo-horiz-rgb-lg-dkbg.gif" alt="7"></a> <ul><li class="last"><a href="http://www.geisauto.com/toyota/"> Curry Toyota...</li></ul> </li>

<img src="/common/logos/nissan/logo-horiz-rgb-lg-dkbg.gif" alt="6"></a> <ul><li class="last"><a href="http://www.geisauto.com/"> Curry Nissan...</li></ul> </li>

Building a Graph

• A graph consists of a fixed set of…– Node Types: {seeds, document, wrapper, mention}– Labeled Directed Edges: {find, derive, extract}

• Each edge asserts that a binary relation r holds• Each edge has an inverse relation r-1 (graph is cyclic)

“ford”, “nissan”, “toyota”

curryauto.com

Wrapper #3

Wrapper #2

Wrapper #1

Wrapper #4

“honda”26.1%

“acura”34.6%

“chevrolet”22.5%

“bmw pittsburgh”8.4%

“volvo chicago”8.4%

find

derive

extract northpointcars.com

Minkov et al. Contextual Search and Name Disambiguation in Email using Graphs. SIGIR 2006

Top three mentions are the seedsTry it out at http://rcwang.com/seal

Relational Set ExpansionSeeds

Additional relevant research

• Alon Halevey and friends:– “Pay as you go”, “on the fly”, data integration (e.g., SIGMOD 98):

integrate partially, then allow user to perform search to make up for inaccuracy of result

• Anhai Doan and friends:– “Best effort” information extraction (SIGMOD 98): write an approximate

program for extraction from web pages, then allow user to perform search to make up for inaccuracy of result

• Semi-structured extensions: – Kushmeric’s ELIXIR (SIGIR 2001); Bernstein’s iSPARQL (eg ESWC

2008)• Soft joins:

– Gravano et al WWW2003: Text Joints in an RDMS for Web Data Integration

– Bayardo et al, WWW2007: Scaling up all-pairs similarity search.– Koudas et al, SIGMOD 2006: Record linkage: similarity measures and

algorithms (survey)

Outline


economic problem


– Classification– Collaborative filtering– Set expansion– Questions?

Documents

Using Machine Learning to Discover and Understand Structured Data William W. Cohen Machine Learning Dept. and Language Technologies Inst. School of Computer