1 Beyond Pages: Supporting Efficient, Scalable Entity Search with Dual-Inversion Index Tao Cheng and Kevin Chang {tcheng3,kcchang}@cs.uiuc.edu Computer.

1

Beyond Pages: Supporting Efficient, Scalable Entity

Search with Dual-Inversion Index

Tao Cheng and Kevin Chang

{tcheng3,kcchang}@cs.uiuc.edu

Computer Science DepartmentUniversity of Illinois at Urbana-Champaign

2

Customer service phone number of Amazon?

Users in Frustration

Search on

Amazon?

Search on

Search Engine?

3

Professors in the area of data mining

Even More Frustration

cs.uiuc.edu cs.uiuc.edu/research

cs.uiuc.edu/research/data

cs.stanford.edu ……

cs.stanford.edu/researchcs.stanford.edu/research/faculty

4

Many many such cases:

The email of Kevin Chang? The papers and presentations of ICDE 2010? Conferences and their due dates on databases in

2010? Sale price of “Canon PowerShot A400”?

Often times, we are looking for data entities, e.g., emails, dates, prices, etc., not pages.

Indeed, according to a recent survey, 52.9% of queries are directly targeting at structured entities [DE Bulletin’09][DE Bulletin’09]: R. Kumar and A. Tomkins, “A Characterization of Online Search Behavior”

Recent Trends: WQAWeb-based Question Answering (WQA)

(Wu 2007, Lin 2003, Brill 2002)

Who is CEO of Dell?

Keywords:“CEO Dell”

Parse Top-k

results

Michael Dell

5

Recent Trends: WIE

6

Specialized Information Extractors

Web Information Extraction (WIE)(Marius 2006, Cafarella 2005, Etzioni 2004)

Pattern: “X is CEO of

Y”

Company CEO

Google Eric Schmidt

IBM S. Palmisano

… …

Recent Trends: TAS

7

Typed-Annotated Search (TAS)(Cheng 2007, Cafarella 2007, Chakrabarti 2006)

Inventor of television?

……

0.60

0.80

0.90

Ranked Entity List

Finding person names nearKeywords “invent” and

“television”

Typed-Annotated Search

8

From Pages to Data EntitiesTraditional Search Entity Search

Keywords Keywords &Entity Type

ResultsResults Support

9

Concretely, what do we mean by Entity Search?

Online Demo.

3TB Corpus of 150M pages 16-machine cluster

24 entity types

10

Entity Search Problem Abstraction

Input: Keywords & Entity Type (optionally with a pattern)

E.g. Amazon Customer Service #phone

Output: Ranked Entity Instances

Ordered by: Score(e)where e is an entity instance ……

0.60

0.80

0.90

Given:

D

Unanimous Requirements across the Trends Context Matching (in document)

Match the target type (say #location) by keywords (e.g., “louvre museum”) that appear in its surrounding context, in certain desired patterns

Global Aggregation (across documents) Match an entity (say, #location = Paris) for as

many times as it appears in numerous pages

11

Computation Challenges

Expensive Context Matching (Join ) Need to perform proximity matching in

documents Beyond simple containment checking

Extensive Global Aggregation (G) Need to perform corpus scale aggregation

A layer that is non-existent in online page retrieval

12

wv L

Traditional Page Retrieval based Approach

13

Who is the CEO of Dell?

Keywords:“CEO Dell”

Analyze top-k

results

Michael Dell

Limitation

• Only top-k documents

• Many random seeks

Our Proposal: Entity-aware Indexing

Inspired by the success of inverted index in enabling efficient IR for searching documents

However, traditional inverted index only aware of keywords and documents How can we make index entity aware?

Our proposal: Dual-Inversion Index Principle I: Document-inverted Index Principle II: Entity-inverted Index

14

Entity-as-keyword: Document-inverted Index

15

2d 12 6d 17

9d 366d 18

9d 34

56d 56 200

257 56d 55 64d 5

68d 56

97d 45

75d 56 97d 47

8p :800-201-7575 10p :408-376-7400

DamazonaD :)(

DservicesD :)(

6d ],23[ 8p 9d],323[ 10p ...],45[ 86p 97d ...],50[ 8p

DphonepD :#)(#

keyword pos

doc id

Document Space Partitioning

2d 12 6d 17

9d 366d 18

9d 34

56d 56 200

257 56d 55 64d 5

68d 56

97d 45

75d 56 97d 47

)(aD

)(sD

6d ],23[ 8p 9d],323[ 10p ...],45[ 86p 97d ...],50[ 8p)(# pD

...200

97d 45

97d 47

97d ],50[ 8p

:)(10 aD

:)(10 sD

:)(#10 pD

Node 10},...,{ 10091

10 ddD

2d 12 6d 17

9d 366d 18

9d 34 257)(1 aD

)(1 sD

6d ],23[ 8p 9d],323[ 10p ],45[ 86p)(#1 pD

Node 1},...,{ 101

1 ddD

Distributed Query Processing over D-inverted Index

17

Join ……

Aggregation

Local

Ranking

},...,{ 1011 ddD },...,{ 10091

10 ddD

Global

wv L Join wv L

SG

)(1 aD )(1 sD )(#1 pD

…

1,8p 1,86p 1,8p

2,8p 1,86p

results, scores

……

)(10 aD )(10 sD )(#10 pD

…

Node 1 Node 10

Entity-as-document: Entity-inverted Index

18

keyword pos entity id entity pos

6d ]17,,23[ 8p 9d ]34,,45[ 86p 97d ]45,,50[ 8p

phoneamazonaE #:)(

6d ]18,,23[ 8p 9d ]36,,45[ 86p 97d ]47,,50[ 8p

phoneservicesE #:)(

Entity Space Partitioning

19

6d ]17,,23[ 8p 9d ]34,,45[ 86p 97d ]45,,50[ 8p)(aE

6d ]18,,23[ 8p 9d ]36,,45[ 86p 97d ]47,,50[ 8p)(sE

...

Node 1

6d ]17,,23[ 8p 97d ]45,,50[ 8p)(1 aE

6d ]18,,23[ 8p 97d ]47,,50[ 8p)(1 sE

},...,{# 1011 ppP Node 9

},...,{# 90819 ppP

)(9 aE

)(9 sE

9d ]34,,45[ 86p

9d ]36,,45[ 86p

...

Distributed Query Processing over E-inverted Index

20

…

Local

RankingGlobal

S

)(1 aE )(1 sE

…

2,8p 1,86p

2,8p 1,86p

results, scores

…

)(9 aE )(9 sE

Node 1 Node 9

},...,{# 1011 ppP },...,{# 90819 ppP

Join

GAggregation

wv L Join

GAggregation

wv L…

…

21

Experiment Setup

Corpus: General crawl of the Web (Aug, 2007), around 3TB with 150M pages.

Entities: 24 diverse entity types

Concrete Applications (Benchmark queries): Yellowpage: #email, #phone, #state, #location, #zipcode CSAcademia: #university, #professor, #research, #email,

#phone

Metrics Used for Evaluation to Measure Throughput & Response Time

Local Processing Time Overall local processing time. Max local processing time

Transfer Time Overall transfer time Max transfer time

Global Processing Time

22

Local Processing Time Comparison

23

Network Transfer Comparison

24

Global Processing Time Comparison

25

Overall Time/Space Summary

26

Generally, ~2 to 4 orders of speedup,

with reasonable space overhead

Dual-Inversion Index

27

Dual-Inversion Index: The two types of indexes can co-exist, and

complement each other

Indexing Configuration

28

Entity Type Level Configuration: Create E-Inverted Index only for popular, space

efficient entities D-Inverted Index for less popular, space

expensive entities

Keyword Level Configuration: Only create E-Inverted Index for <keyword,

entity> pairs, when they are related, e.g., queried often from query log

Conclusion

Identify essential computation requirements for entity search

Dual-inversion indexing and partition schemes for efficient and scalable query processing Document-inverted index Entity-inverted index

Verify over large-scale corpus with real applications

29

30

Thanks much for coming!

Questions?

TopK Convergence

31

References of Related Work Index Design

Junghoo Cho and Sridhar Rajagopalan. A fast regular expression indexing engine. In ICDE, 2002.

Hugh E. Williams, Justin Zobel, and Dirk Bahle. Fast phrase querying with combined indexes. ACM Trans. Inf. Syst., 22(4):573–594, 2004.

Xiaohui Long and Torsten Suel. Three-level caching for efficient query processing in large web search engines. In WWW, 2005.

Michael Cafarella and Oren Etzioni. A search engine for large-corpus language applications. In WWW, 2005.

Question Answering S. Abney, M. Collins, and A. Singhal. Answer extraction. In ANLP, 2000. E. Brill, S. Dumais, and M. Banko. An analysis of the askmsr question-answering

system. In EMNLP, 2002. Cody C. T. Kwok, Oren Etzioni, and Daniel S. Weld. Scaling question answering

to the web. In WWW, 2001. Jimmy J. Lin and Boris Katz. Question answering from the web using knowledge

annotation and knowledge mining techniques. In CIKM, 2003.

32

Search Interface

33

Query I: Amazon Customer Service Phone

34

Results# of Supporting Page

Representative Supporting Pages

Query II: Professors in Data Mining

35

Query III: University of California Locations

36