AnHai Doan, Pedro Domingos, Alon Halevy

1
Reconciling Schemas of Disparate Data Sources A
Machine Learning Approach
The LSD Project

• AnHai Doan, Pedro Domingos, Alon Halevy
• University of Washington

2
Data Integration
Find houses with four bathrooms priced under
500,000
mediated schema
source schema 2
source schema 3
source schema 1
homes.com
realestate.com
homeseekers.com
3
Semantic Mappings between Schemas

• Mediated source schemas XML DTDs

house
address
num-baths
contact-info
agent-name agent-phone
1-1 mapping
non 1-1 mapping
house
location contact
full-baths
half-baths
name phone
4
Current State of Affairs

• Finding semantic mappings is now the bottleneck!
• largely done by hand
• labor intensive error prone
• Will only be exacerbated
• data sharing XML become pervasive
• proliferation of DTDs
• translation of legacy data
• reconciling ontologies on the semantic web
• Need (semi-)automatic approaches to scale up!

5
The LSD (Learning Source Descriptions) Approach

• Suppose user wants to integrate 100 data sources
• 1. User
• manually creates mappings for a few sources, say
  3
• shows LSD these mappings
• 2. LSD learns from the mappings
• 3. LSD proposes mappings for remaining 97 sources

6
Example
Mediated schema
address price agent-phone
description
location listed-price phone
comments
Learned hypotheses
Schema of realestate.com
If phone occurs in the name gt agent-phone
listed-price 250,000 110,000 ...
location Miami, FL Boston, MA ...
phone [phone] [phone] ...
comments Fantastic house Great location ...
realestate.com
If fantastic great occur frequently in
data values gt description
homes.com
price 550,000 320,000 ...
contact-phone [phone] [phone] ...
extra-info Beautiful yard Great beach ...
7
Our Contributions

• 1. Use of multi-strategy learning
• well-suited to exploit multiple types of
  knowledge
• highly modular extensible
• 2. Extend learning to incorporate constraints
• handle a wide range of domain user-specified
  constraints
• 3. Develop XML learner
• exploit hierarchical nature of XML

8
Multi-Strategy Learning

• Use a set of base learners
• each exploits well certain types of information
• Match schema elements of a new source
• apply the base learners
• combine their predictions using a meta-learner
• Meta-learner
• uses training sources to measure base learner
  accuracy
• weighs each learner based on its accuracy

9
Base Learners

• Input
• schema information name, proximity, structure,
  ...
• data information value, format, ...
• Output
• prediction weighted by confidence score
• Examples
• Name learner
• agent-name gt (name,0.7), (phone,0.3)
• Naive Bayes learner
• Kent, WA gt (address,0.8),
  (name,0.2)
• Great location gt (description,0.9),
  (address,0.1)

10
Training the Learners
Mediated schema
address price agent-phone
description
location listed-price phone
comments
Schema of realestate.com
Name Learner
(location, address) (listed-price, price) (phone,
agent-phone) (comments, description) ...
ltlocationgt Miami, FL lt/gt ltlisted-pricegt
250,000lt/gt ltphonegt [phone]lt/gt
ltcommentsgt Fantastic house lt/gt
realestate.com
Naive Bayes Learner
ltlocationgt Boston, MA lt/gt ltlisted-pricegt
110,000lt/gt ltphonegt [phone]lt/gt
ltcommentsgt Great location lt/gt
(Miami, FL, address) ( 250,000,
price) ([phone], agent-phone) (Fantastic
house, description) ...
11
Applying the Learners
Mediated schema
Schema of homes.com
address price agent-phone
description
area day-phone extra-info
Name Learner Naive Bayes
ltareagtSeattle, WAlt/gt ltareagtKent,
WAlt/gt ltareagtAustin, TXlt/gt
(address,0.8), (description,0.2) (address,0.6),
(description,0.4) (address,0.7), (description,0.3)
Meta-Learner
Name Learner Naive Bayes
Meta-Learner
(address,0.7), (description,0.3)
ltday-phonegt(278) 345 7215lt/gt ltday-phonegt(617) 335
2315lt/gt ltday-phonegt(512) 427 1115lt/gt
(agent-phone,0.9), (description,0.1)
(address,0.6), (description,0.4)
ltextra-infogtBeautiful yardlt/gt ltextra-infogtGreat
beachlt/gt ltextra-infogtClose to Seattlelt/gt
12
Domain Constraints

• Impose semantic regularities on sources
• verified using schema or data
• Examples
• a address b address a b
• a house-id a is a key
• a agent-info b agent-name b is
  nested in a
• Can be specified up front
• when creating mediated schema
• independent of any actual source schema

13
The Constraint Handler
Domain Constraints a address b adderss
a b
Predictions from Meta-Learner
area (address,0.7),
(description,0.3) contact-phone
(agent-phone,0.9), (description,0.1) extra-info
(address,0.6), (description,0.4)
0.3 0.1 0.4 0.012
area address contact-phone
agent-phone extra-info description
area address contact-phone
agent-phone extra-info address
0.7 0.9 0.6 0.378
0.7 0.9 0.4 0.252

• Can specify arbitrary constraints
• User feedback domain constraint
• ad-id house-id
• Extended to handle domain heuristics
• a agent-phone b agent-name a b are
  usually close to each other

14
Putting It All Together the LSD System
Matching Phase
Training Phase
Mediated schema
Source schemas
Domain Constraints
Data listings
Training data for base learners
User Feedback
Constraint Handler
L1
L2
Lk
Mapping Combination

• Base learners Name Learner, XML learner, Naive
  Bayes, Whirl learner
• Meta-learner
• uses stacking TingWitten99, Wolpert92
• returns linear weighted combination of base
  learners predictions

15
Empirical Evaluation

• Four domains
• Real Estate I II, Course Offerings, Faculty
  Listings
• For each domain
• create mediated DTD domain constraints
• choose five sources
• extract convert data listings into XML
• mediated DTDs 14 - 66 elements, source DTDs 13
  - 48

• Ten runs for each experiment - in each run
• manually provide 1-1 mappings for 3 sources
• ask LSD to propose mappings for remaining 2
  sources
• accuracy of 1-1 mappings correctly identified

16
High Matching Accuracy
Average Matching Acccuracy ()
LSDs accuracy 71 - 92
Best single base learner 42 - 72
Meta-learner 5 - 22
Constraint handler 7 - 13 XML
learner 0.8 - 6
17
Performance Sensitivity
Average matching accuracy ()
Number of data listings per source
18
Contribution of Schema vs. Data
Average matching accuracy ()

• More experiments in the paper!

19
Related Work

• Rule-based approaches
• TRANSCM MiloZohar98, ARTEMIS
  CastanoAntonellis99, Palopoli et. al. 98,
  CUPID Madhavan et. al. 01
• utilize only schema information
• Learner-based approaches
• SEMINT LiClifton94, ILA PerkowitzEtzioni95
• employ a single learner, limited applicability
• Others
• DELTA Clifton et. al. 97, CLIO Miller et. al.
  00Yan et. al. 01
• Multi-strategy learning in other domains
• series of workshops 91,93,96,98,00
• Freitag98, Proverb Keim et. al. 99

20
Summary

• LSD project
• applies machine learning to schema matching
• Main ideas contributions
• use of multi-strategy learning
• extend learning to handle domain user-specified
  constraints
• develop XML learner
• System design A contribution to generic
  schema-matching
• highly modular extensible
• handle multiple types of knowledge
• continuously improve over time

21
Ongoing Future Work

• Improve accuracy
• address current system limitations
• Extend LSD to more complex mappings
• Apply LSD to other application contexts
• data translation
• data warehousing
• e-commerce
• information extraction
• semantic web
• www.cs.washington.edu/homes/anhai/lsd.h
  tml

22
Contribution of Each Component
Average Matching Acccuracy ()
Without Name Learner Without Naive Bayes Without
Whirl Learner Without Constraint Handler The
complete LSD system
23
Exploiting Hierarchical Structure

• Existing learners flatten out all structures
• Developed XML learner
• similar to the Naive Bayes learner
• input instance bag of tokens
• differs in one crucial aspect
• consider not only text tokens, but also structure
  tokens

ltcontactgt ltnamegt Gail Murphy lt/namegt ltfirmgt
MAX Realtors lt/firmgt lt/contactgt
ltdescriptiongt Victorian house with a view.
Name your price! To see it, contact Gail
Murphy at MAX Realtors. lt/descriptiongt