Evaluation of IR Systems

1
Evaluation of IR Systems

• Adapted from Lectures by
• Prabhakar Raghavan (Yahoo and Stanford) and
  Christopher Manning (Stanford)

2
This lecture

• How do we summarize results?
• Making our good results usable to a user
• How do we know if our results are any good?
• Necessary to determine effectiveness of
• Ranking function (dot-product, cosine, )
• Term selection (stopword removal, stemming)
• Term weighting (TF, TF-IDF,)
• How far down the ranked list will a user need to
  look to find some/all relevant documents?
• Evaluating a search engine
• Benchmarks
• Precision and recall

3
Results summaries
4
Summaries

• Having ranked the documents matching a query, we
  wish to present a results list (answer set)
• Most commonly, the document title plus a short
  summary
• The title is typically automatically extracted
  from document metadata.
• What about the summaries?

5
Summaries

• Two basic kinds
• Static (cf. extractive)
• Dynamic (cf. abstractive)
• A static summary of a document is always the
  same, regardless of the query that hit the doc.
• Dynamic summaries are query-dependent and attempt
  to explain why the document was retrieved for the
  query at hand.

6
Static summaries

• In typical systems, the static summary is a
  subset of the document.
• Simplest heuristic the first 50 (or so this
  can be varied) words of the document
• Summary cached at indexing time
• More sophisticated extract from each document a
  set of key sentences
• Simple NLP heuristics to score each sentence.
• Summary is made up of top-scoring sentences.
• Most sophisticated NLP used to synthesize a
  summary
• Seldom used in IR (cf. text summarization work)

7
Dynamic summaries

• Present one or more windows within the document
  that contain several of the query terms
• KWIC snippets Keyword in Context presentation
• Generated in conjunction with scoring
• If query found as a phrase, the/some occurrences
  of the phrase in the doc
• If not, windows within the doc that contain
  multiple query terms
• The summary itself gives the entire content of
  the window all terms, not only the query terms.

8
Generating dynamic summaries

• If we have only a positional index, we cannot
  (easily) reconstruct context surrounding hits
• If we cache the documents at index time, can run
  the window through it, cueing to hits found in
  the positional index
• E.g., positional index says the query is a
  phrase in position 4378 so we go to this
  position in the cached document and stream out
  the content
• Most often, cache a fixed-size prefix of the doc

9
Dynamic summaries

• Producing good dynamic summaries is a tricky
  optimization problem
• The real estate for the summary is normally small
  and fixed
• Want short item, so show as many KWIC matches as
  possible, and perhaps other things like title
• Want snippets to be long enough to be useful
• Want linguistically well-formed snippets users
  prefer snippets that contain complete phrases
• Want snippets maximally informative about doc
• But users really like snippets, even if they
  complicate IR system design

10
Evaluating search engines
11
Measures for a search engine

• How fast does it index
• Number of documents/hour
• (Average document size)
• How fast does it search
• Latency as a function of index size
• Expressiveness of query language
• Ability to express complex information needs
• Speed on complex queries

12
Measures for a search engine

• All of the preceding criteria are measurable we
  can quantify speed/size we can make
  expressiveness precise
• The key measure user happiness
• What is this?
• Speed of response/size of index are factors
• But blindingly fast, useless answers wont make a
  user happy
• Need a way of quantifying user happiness

13
Data Retrieval vs Information Retrieval

• DR Performance Evaluation (after establishing
  correctness)
• Response time
• Index space
• IR Performance Evaluation
• How relevant is the answer set? (required to
  establish functional correctness, e.g., through
  benchmarks)

14
Measuring user happiness

• Issue who is the user we are trying to make
  happy?
• Web engine user finds what they want and return
  to the engine
• Can measure rate of return users
• eCommerce site user finds what they want and
  make a purchase
• Is it the end-user, or the eCommerce site, whose
  happiness we measure?
• Measure time to purchase, or fraction of
  searchers who become buyers?

15
Measuring user happiness

• Enterprise (company/govt/academic) Care about
  user productivity
• How much time do my users save when looking for
  information?
• Many other criteria having to do with breadth of
  access, secure access, etc.

16
Happiness elusive to measure

• Commonest proxy relevance of search results
• But how do you measure relevance?
• We will detail a methodology here, then examine
  its issues
• Relevance measurement requires 3 elements
• A benchmark document collection
• A benchmark suite of queries
• A binary assessment of either Relevant or
  Irrelevant for each query-doc pair
• Some work on more-than-binary, but not the
  standard

17
Evaluating an IR system

• Relevance is assessed relative to the information
  need, not the query
• E.g., Information need I'm looking for
  information on whether drinking red wine is more
  effective at reducing your risk of heart attacks
  than white wine.
• Query wine red white heart attack effective
• You evaluate whether the doc addresses the
  information need, not whether it has those words

18
Difficulties with gauging Relevancy

• Relevancy, from a human standpoint, is
• Subjective Depends upon a specific users
  judgment.
• Situational Relates to users current needs.
• Cognitive Depends on human perception and
  behavior.
• Dynamic Changes over time.

19
Standard relevance benchmarks

• TREC - National Institute of Standards and
  Testing (NIST) has run a large IR test bed for
  many years
• Reuters and other benchmark doc collections used
• Retrieval tasks specified
• sometimes as queries
• Human experts mark, for each query and for each
  doc, Relevant or Irrelevant
• or at least for subset of docs that some system
  returned for that query

20
Unranked retrieval evaluationPrecision and
Recall

• Precision fraction of retrieved docs that are
  relevant P(relevantretrieved)
• Recall fraction of relevant docs that are
  retrieved P(retrievedrelevant)
• Precision P tp/(tp fp)
• Recall R tp/(tp fn)

Relevant Not Relevant
Retrieved tp fp
Not Retrieved fn tn
21
Precision and Recall in Practice

• Precision
• The ability to retrieve top-ranked documents that
  are mostly relevant.
• The fraction of the retrieved documents that is
  relevant.
• Recall
• The ability of the search to find all of the
  relevant items in the corpus.
• The fraction of the relevant documents that is
  retrieved.

22
Precision and Recall
23
Computing Recall/Precision Points An Example
Let total of relevant docs 6 Check each new
recall point
R1/60.167 P1/11
R2/60.333 P2/21
R3/60.5 P3/40.75
R4/60.667 P4/60.667
Missing one relevant document. Never reach 100
recall
R5/60.833 p5/130.38
24
Accuracy

• Given a query, an engine classifies each doc as
  Relevant or Irrelevant.
• Accuracy of an engine the fraction of these
  classifications that is correct.
• Why is this not a very useful evaluation measure
  in IR?

25
Why not just use accuracy?

• How to build a 99.9999 accurate search engine on
  a low budget.
• People doing information retrieval want to find
  something and have a certain tolerance for junk.

Snoogle.com
Search for
0 matching results found.
26
Precision/Recall

• You can get high recall (but low precision) by
  retrieving all docs for all queries!
• Recall is a non-decreasing function of the number
  of docs retrieved
• In a good system, precision decreases as either
  number of docs retrieved or recall increases
• A fact with strong empirical confirmation

27
Trade-offs
1
Precision
0
1
Recall
28
Difficulties in using precision/recall

• Should average over large corpus/query ensembles
• Need human relevance assessments
• People arent reliable assessors
• Assessments have to be binary
• Nuanced assessments?
• Heavily skewed by corpus/authorship
• Results may not translate from one domain to
  another

29
A combined measure F

• Combined measure that assesses this tradeoff is F
  measure (weighted harmonic mean)
• People usually use balanced F1 measure
• i.e., with ? 1 or ? ½
• Harmonic mean is a conservative average
• See CJ van Rijsbergen, Information Retrieval

30
F1 and other averages
31
Evaluating ranked results

• Evaluation of ranked results
• The system can return any number of results
• By taking various numbers of the top returned
  documents (levels of recall), the evaluator can
  produce a precision-recall curve

32
A precision-recall curve
33
Evaluation

• Graphs are good, but people want summary
  measures!
• Precision at fixed retrieval level
• Perhaps most appropriate for web search all
  people want are good matches on the first one or
  two results pages
• But has an arbitrary parameter of k
• 11-point interpolated average precision
• The standard measure in the TREC competitions
  you take the precision at 11 levels of recall
  varying from 0 to 1 by tenths of the documents,
  using interpolation (the value for 0 is always
  interpolated!), and average them
• Evaluates performance at all recall levels

34
Typical (good) 11 point precisions

• SabIR/Cornell 8A1 11pt precision from TREC 8
  (1999)

35
11 point precisions
36
Creating Test Collectionsfor IR Evaluation
37
Test Corpora
38
From corpora to test collections

• Still need
• Test queries
• Relevance assessments
• Test queries
• Must be germane to docs available
• Best designed by domain experts
• Random query terms generally not a good idea
• Relevance assessments
• Human judges, time-consuming
• Are human panels perfect?

39
Unit of Evaluation

• We can compute precision, recall, F, and ROC
  curve for different units.
• Possible units
• Documents (most common)
• Facts (used in some TREC evaluations)
• Entities (e.g., car companies)
• May produce different results. Why?

40
Can we avoid human judgment?

• Not really
• Makes experimental work hard
• Especially on a large scale
• In some very specific settings, can use proxies
• Example below, approximate vector space retrieval
• But once we have test collections, we can reuse
  them (so long as we dont overtrain too badly)

41
Approximate vector retrieval

• Given n document vectors and a query, find the k
  doc vectors closest to the query.
• Exact retrieval we know of no better way than
  to compute cosines from the query to every doc
• Approximate retrieval schemes such as cluster
  pruning
• Given such an approximate retrieval scheme, how
  do we measure its goodness?

42
Approximate vector retrieval

• Let G(q) be the ground truth of the actual k
  closest docs on query q
• Let A(q) be the k docs returned by approximate
  algorithm A on query q
• For performance we would measure A(q) ? G(q)
• Is this the right measure?

43
Alternative proposal

• Focus instead on how A(q) compares to G(q).
• Goodness can be measured here in cosine proximity
  to q we sum up q?d over d? A(q).
• Compare this to the sum of q?d over d? G(q).
• Yields a measure of the relative goodness of A
  vis-à-vis G.
• Thus A may be 90 as good as the ground-truth
  G, without finding 90 of the docs in G.
• For scored retrieval, this may be acceptable
• Most web engines dont always return the same
  answers for a given query.

44
Other Evaluation Measures andTREC Benchmarks
Adapted from Slides Attributed to Prof. Dik Lee
(Univ. of Science and Tech, Hong Kong)
45
R- Precision

• Precision at the R-th position in the ranking of
  results for a query that has R relevant documents.

R of relevant docs 6
R-Precision 4/6 0.67
46
E Measure (parameterized F Measure)

• A variant of F measure that allows weighting
  emphasis on precision over recall
• Value of ? controls trade-off
• ? 1 Equally weight precision and recall (EF).
• ? gt 1 Weight precision more.
• ? lt 1 Weight recall more.

47
Fallout Rate

• Problems with both precision and recall
• Number of irrelevant documents in the collection
  is not taken into account.
• Recall is undefined when there is no relevant
  document in the collection.
• Precision is undefined when no document is
  retrieved.

48
Subjective Relevance Measure

• Novelty Ratio The proportion of items retrieved
  and judged relevant by the user and of which they
  were previously unaware.
• Ability to find new information on a topic.
• Coverage Ratio The proportion of relevant items
  retrieved out of the total relevant documents
  known to a user prior to the search.
• Relevant when the user wants to locate documents
  which they have seen before (e.g., the budget
  report for Year 2000).

49
Other Factors to Consider

• User effort Work required from the user in
  formulating queries, conducting the search, and
  screening the output.
• Response time Time interval between receipt of a
  user query and the presentation of system
  responses.
• Form of presentation Influence of search output
  format on the users ability to utilize the
  retrieved materials.
• Collection coverage Extent to which any/all
  relevant items are included in the document
  corpus.

50
Early Test Collections

• Previous experiments were based on the SMART
  collection which is fairly small.
  (ftp//ftp.cs.cornell.edu/pub/smart)
• Collection Number Of Number Of Raw Size
• Name Documents Queries (Mbytes)
• CACM 3,204 64 1.5
• CISI 1,460 112 1.3
• CRAN 1,400 225 1.6
• MED 1,033 30 1.1
• TIME 425 83 1.5
• Different researchers used different test
  collections and evaluation techniques.

51
The TREC Benchmark

• TREC Text REtrieval Conference
  (http//trec.nist.gov/)
• Originated from the TIPSTER program sponsored
  by
• Defense Advanced Research Projects Agency
  (DARPA).
• Became an annual conference in 1992,
  co-sponsored by the
• National Institute of Standards and Technology
  (NIST) and
• DARPA.
• Participants are given parts of a standard set
  of documents
• and TOPICS (from which queries have to be
  derived) in
• different stages for training and testing.
• Participants submit the P/R values for the final
  document
• and query corpus and present their results at
  the conference.

52
The TREC Objectives

• Provide a common ground for comparing different
  IR
• techniques.
• Same set of documents and queries, and same
  evaluation method.
• Sharing of resources and experiences in
  developing the
• benchmark.
• With major sponsorship from government to develop
  large benchmark collections.
• Encourage participation from industry and
  academia.
• Development of new evaluation techniques,
  particularly for
• new applications.
• Retrieval, routing/filtering, non-English
  collection, web-based collection, question
  answering.

53
TREC Advantages

• Large scale (compared to a few MB in the SMART
  Collection).
• Relevance judgments provided.
• Under continuous development with support from
  the U.S. Government.
• Wide participation
• TREC 1 28 papers 360 pages.
• TREC 4 37 papers 560 pages.
• TREC 7 61 papers 600 pages.
• TREC 8 74 papers.

54
TREC Tasks

• Ad hoc New questions are being asked on a static
  set of data.
• Routing Same questions are being asked, but new
  information is being searched. (news clipping,
  library profiling).
• New tasks added after TREC 5 - Interactive,
  multilingual, natural language, multiple database
  merging, filtering, very large corpus (20 GB, 7.5
  million documents), question answering.

55
Characteristics of the TREC Collection

• Both long and short documents (from a few hundred
  to over one thousand unique terms in a document).
• Test documents consist of
• WSJ Wall Street Journal articles (1986-1992)
  550 M
• AP Associate Press Newswire (1989)
  514 M
• ZIFF Computer Select Disks (Ziff-Davis
  Publishing) 493 M
• FR Federal Register 469 M
• DOE Abstracts from Department of Energy
  reports 190 M

56
More Details on Document Collections

• Volume 1 (Mar 1994) - Wall Street Journal (1987,
  1988, 1989), Federal Register (1989), Associated
  Press (1989), Department of Energy abstracts, and
  Information from the Computer Select disks (1989,
  1990)
• Volume 2 (Mar 1994) - Wall Street Journal (1990,
  1991, 1992), the Federal Register (1988),
  Associated Press (1988) and Information from the
  Computer Select disks (1989, 1990)
• Volume 3 (Mar 1994) - San Jose Mercury News
  (1991), the Associated Press (1990), U.S. Patents
  (1983-1991), and Information from the Computer
  Select disks (1991, 1992)
• Volume 4 (May 1996) - Financial Times Limited
  (1991, 1992, 1993, 1994), the Congressional
  Record of the 103rd Congress (1993), and the
  Federal Register (1994).
• Volume 5 (Apr 1997) - Foreign Broadcast
  Information Service (1996) and the Los Angeles
  Times (1989, 1990).

57
TREC Disk 4,5
58
Sample Document (with SGML)

• ltDOCgt
• ltDOCNOgt WSJ870324-0001 lt/DOCNOgt
• ltHLgt John Blair Is Near Accord To Sell Unit,
  Sources Say lt/HLgt
• ltDDgt 03/24/87lt/DDgt
• ltSOgt WALL STREET JOURNAL (J) lt/SOgt
• ltINgt REL TENDER OFFERS, MERGERS, ACQUISITIONS
  (TNM) MARKETING, ADVERTISING (MKT)
  TELECOMMUNICATIONS, BROADCASTING, TELEPHONE,
  TELEGRAPH (TEL) lt/INgt
• ltDATELINEgt NEW YORK lt/DATELINEgt
• ltTEXTgt
• John Blair amp Co. is close to an
  agreement to sell its TV station advertising
  representation operation and program production
  unit to an investor group led by James H.
  Rosenfield, a former CBS Inc. executive, industry
  sources said. Industry sources put the value of
  the proposed acquisition at more than 100
  million. ...
• lt/TEXTgt
• lt/DOCgt

59
Sample Query (with SGML)

• lttopgt
• ltheadgt Tipster Topic Description
• ltnumgt Number 066
• ltdomgt Domain Science and Technology
• lttitlegt Topic Natural Language Processing
• ltdescgt Description Document will identify a type
  of natural language processing technology which
  is being developed or marketed in the U.S.
• ltnarrgt Narrative A relevant document will
  identify a company or institution developing or
  marketing a natural language processing
  technology, identify the technology, and identify
  one of more features of the company's product.
• ltcongt Concept(s) 1. natural language processing
  2. translation, language, dictionary
• ltfacgt Factor(s)
• ltnatgt Nationality U.S.lt/natgt
• lt/facgt
• ltdefgt Definitions(s)
• lt/topgt

60
TREC Properties

• Both documents and queries contain many different
  kinds of information (fields).
• Generation of the formal queries (Boolean, Vector
  Space, etc.) is the responsibility of the system.
• A system may be very good at querying and
  ranking, but if it generates poor queries from
  the topic, its final P/R would be poor.

61
Two more TREC Document Examples
62
Another Example of TREC Topic/Query
63
Evaluation

• Summary table statistics Number of topics,
  number of documents retrieved, number of relevant
  documents.
• Recall-precision average Average precision at 11
  recall levels (0 to 1 at 0.1 increments).
• Document level average Average precision when 5,
  10, .., 100, 1000 documents are retrieved.
• Average precision histogram Difference of the
  R-precision for each topic and the average
  R-precision of all systems for that topic.

64
(No Transcript)
65
Cystic Fibrosis (CF) Collection

• 1,239 abstracts of medical journal articles on
  CF.
• 100 information requests (queries) in the form of
  complete English questions.
• Relevant documents determined and rated by 4
  separate medical experts on 0-2 scale
• 0 Not relevant.
• 1 Marginally relevant.
• 2 Highly relevant.

66
CF Document Fields

• MEDLINE access number
• Author
• Title
• Source
• Major subjects
• Minor subjects
• Abstract (or extract)
• References to other documents
• Citations to this document

67
Sample CF Document
AN 74154352 AU Burnell-R-H. Robertson-E-F. TI
Cystic fibrosis in a patient with Kartagener
syndrome. SO Am-J-Dis-Child. 1974 May. 127(5). P
746-7. MJ CYSTIC-FIBROSIS co. KARTAGENER-TRIAD
co. MN CASE-REPORT. CHLORIDES an. HUMAN.
INFANT. LUNG ra. MALE. SITUS-INVERSUS co,
ra. SODIUM an. SWEAT an. AB A patient
exhibited the features of both Kartagener
syndrome and cystic fibrosis. At most, to the
authors' knowledge, this represents the third
such report of the combination. Cystic
fibrosis should be excluded before a diagnosis of
Kartagener syndrome is made. RF 001
KARTAGENER M BEITR KLIN TUBERK
83 489 933 002 SCHWARZ V
ARCH DIS CHILD 43 695 968
003 MACE JW CLIN PEDIATR
10 285 971 CT 1 BOCHKOVA DN
GENETIKA (SOVIET GENETICS) 11 154
975 2 WOOD RE AM REV RESPIR
DIS 113 833 976 3 MOSSBERG
B MT SINAI J MED 44
837 977
68
Sample CF Queries
QN 00002 QU Can one distinguish between the
effects of mucus hypersecretion and infection
on the submucosal glands of the respiratory tract
in CF? NR 00007 RD 169 1000 434 1001 454 0100
498 1000 499 1000 592 0002 875 1011 QN
00004 QU What is the lipid composition of CF
respiratory secretions? NR 00009 RD 503 0001
538 0100 539 0100 540 0100 553 0001 604 2222
669 1010 711 2122 876 2222
NR Number of Relevant documents RD Relevant
Documents Ratings code Four 0-2 ratings, one
from each expert