Evaluation of IR Systems

1. Evaluation of IR Systems Adapted from Lectures by Prabhakar Raghavan (Yahoo and Stanford) and Christopher Manning (Stanford) L08Evaluation

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 won’t 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 thequery • 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 user’s judgment. • Situational: Relates to user’s 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 evaluation:Precision and Recall • Precision: fraction of retrieved docs that are relevant = P(relevant|retrieved) • Recall: fraction of relevant docs that are retrieved = P(retrieved|relevant) • Precision P = tp/(tp + fp) • Recall R = tp/(tp + fn)

21. Precision and Recall in Practice • Precision • The ability to retrievetop-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: R=1/6=0.167; P=1/1=1 R=2/6=0.333; P=2/2=1 R=3/6=0.5; P=3/4=0.75 R=4/6=0.667; P=4/6=0.667 Missing one relevant document. Never reach 100% recall R=5/6=0.833; p=5/13=0.38

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 findsomething 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. Returns relevant documents but misses many useful ones too The ideal Returns most relevant documents but includes lot of junk 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 aren’t 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 F1measure • 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 don’t 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 don’t 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 (E=F). •  > 1: Weight precision more. •  < 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 user’s 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.