Schema Matching in a Large Scale Schema Matching in a Large Scale - - PowerPoint PPT Presentation

Schema Matching in a Large Scale Schema Matching in a Large Scale Personal Schema Based Querying Personal Schema Based Querying

Marko Smiljani, Maurice van Keulen, Willem Jonker

Dutch Dutch-Belgian Database Day Belgian Database Day - December 3, 2004 December 3, 2004 - Antwerp, Belgium Antwerp, Belgium

in this talk in this talk

• motivation

motivation personal schema based querying

• understanding

understanding formalizing the schema matching problem

• solving

solving clustering in schema matching

• validating

validating semantic validation without semantics

motivation motivation

data

mediated schema mediated schema

data data

//account[number=1234]/owner

mediator

data

personal schema personal schema

data data

//account[number=1234]/owner

PSQ

PSQ – Personal Schema Based Query Answering System

• data

schema repository schema loader

• select
• architecture

architecture

schemas

• D

Dé éj jà à Vu Vu

goals and issues goals and issues

goals

• efficiency of schema matching

(time-to-last, time-to-first)

• effectiveness of schema matching

(precision/recall) issues

• trees vs. graphs
• the objective function

understanding understanding

schema matching schema matching

hints

well known framework, well known framework,

• ffering a range of approaches for efficient problem solving
• ffering a range of approaches for efficient problem solving

formalism formalism

constraint optimization problem constraint optimization problem

formalism formalism

ranking correctness

finding a solution finding a solution

the idea of clustering the idea of clustering

distance based clustering

why clustering? why clustering?

• clusters can be ranked
• search space is reduced

clustering approaches (and issues) clustering approaches (and issues) k-medoid

• how to initialize
• pre-computation of distance

hand made linear-time clustering

• make it intelligent,

yet keep it close to linear-time

• clustering method has to be scalable

validation validation

validation paradox validation paradox

• semantic validation

semantic validation does not like large search spaces! does not like large search spaces!

vs.

• vs. .
• clustering is only useful in large search spaces!

clustering is only useful in large search spaces!

H A T P = T / A R = T / H

s e a r c h s p a c e s e a r c h s p a c e

estimating the precision and recall estimating the precision and recall

• size based
• order based

H A T P = T / A R = T / H H

T

B

size based quality estimation size based quality estimation

R12 = B / A

n

• c

l u s t e r i n g y e s c l u s t e r i n g

B

size based quality estimation size based quality estimation

NO CLUSTERING NO CLUSTERING

• CLUST. BEST CASE
• CLUST. BEST CASE
• CLUST. WORST CASE
• CLUST. WORST CASE

B/A = 93% B

H

• rder based quality estimation
• rder based quality estimation
• ✁

n

• c

l u s t e r i n g n

• c

l u s t e r i n g y e s c l u s t e r i n g y e s c l u s t e r i n g

• rder based quality estimation
• rder based quality estimation

NO CLUSTERING NO CLUSTERING

• CLUST. ALG 1
• CLUST. ALG 1
• CLUST. ALG 2
• CLUST. ALG 2

what comes next what comes next

• add intelligence to clustering
• impact of other hints on clustering
• using graphs

En dat was het dan! En dat was het dan!

Vragen? Vragen?