Information Retrieval
CSE 8337 (Part D)Spring 2009
Some Material for these slides obtained from:Modern Information Retrieval by Ricardo Baeza-Yates and Berthier Ribeiro-Neto
http://www.sims.berkeley.edu/~hearst/irbook/Data Mining Introductory and Advanced Topics by Margaret H. Dunham
http://www.engr.smu.edu/~mhd/bookIntroduction to Information Retrieval by Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schutze
http://informationretrieval.org
CSE 8337 Spring 2009 2
CSE 8337 Outline• Introduction• Simple Text Processing• Boolean Queries• Web Searching/Crawling• Indexes• Vector Space Model• Matching• Evaluation
CSE 8337 Spring 2009 3
Why System Evaluation? There are many retrieval models/
algorithms/ systems, which one is the best?
What does best mean? IR evaluation may not actually look at
traditional CS metrics of space/time. What is the best component for:
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?
CSE 8337 Spring 2009 4
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
Uncluttered UI Is it free?
CSE 8337 Spring 2009 5
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
CSE 8337 Spring 2009 6
Happiness: elusive to measure Most common proxy: relevance of
search results But how do you measure relevance? We will detail a methodology here, then
examine its issues Relevant measurement requires 3
elements:1. A benchmark document collection2. A benchmark suite of queries3. A usually binary assessment of either
Relevant or Nonrelevant for each query and each document
CSE 8337 Spring 2009 7
Difficulties in Evaluating IR Systems
Effectiveness is related to the relevancy of retrieved items.
Relevancy is not typically binary but continuous.
Even if relevancy is binary, it can be a difficult judgment to make.
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.
CSE 8337 Spring 2009 8
How to perform evaluation
Start with a corpus of documents. Collect a set of queries for this corpus. Have one or more human experts
exhaustively label the relevant documents for each query.
Typically assumes binary relevance judgments.
Requires considerable human effort for large document/query corpora.
CSE 8337 Spring 2009 9
IR Evaluation Metrics Precision/Recall
P/R graph Regular Smoothing
Interpolating Averaging
ROC Curve MAP R-Precision P/R points
F-Measure E-Measure Fallout Novelty Coverage Utility ….
CSE 8337 Spring 2009 10
documents relevant of number Totalretrieved documents relevant of Number recall
retrieved documents of number Totalretrieved documents relevant of Number precision
Relevant documents
Retrieved documents
Entire document collection
retrieved & relevant
not retrieved but relevant
retrieved & irrelevant
Not retrieved & irrelevant
retrieved not retrieved
rele
vant
irrel
evan
t
Precision and Recall
CSE 8337 Spring 2009 11
Precision and Recall Precision
The ability to retrieve top-ranked documents that are mostly relevant.
Recall The ability of the search to find all of
the relevant items in the corpus.
CSE 8337 Spring 2009 12
Determining Recall is Difficult Total number of relevant items is
sometimes not available: Sample across the database and
perform relevance judgment on these items.
Apply different retrieval algorithms to the same database for the same query. The aggregate of relevant items is taken as the total relevant set.
CSE 8337 Spring 2009 13
Trade-off between Recall and Precision
10
1
Recall
Prec
isio
n
The ideal
Desired areas
Returns relevant documents butmisses many useful ones too
Returns most relevantdocuments but includes lots of junk
CSE 8337 Spring 2009 14
Recall-Precision Graph Example
Recall-Precision Graph
00.20.40.60.8
1
0 0.5 1Recall
Prec
ision
CSE 8337 Spring 2009 15
A precision-recall curve
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.2 0.4 0.6 0.8 1.0
Pre
cisi
on
Recall
CSE 8337 Spring 2009 16
Recall-Precision Graph Smoothing Avoid sawtooth lines by smoothing Interpolate for one query Average across queries
CSE 8337 Spring 2009 17
Interpolating a Recall/Precision Curve
Interpolate a precision value for each standard recall level: rj {0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9,
1.0} r0 = 0.0, r1 = 0.1, …, r10=1.0
The interpolated precision at the j-th standard recall level is the maximum known precision at any recall level between the j-th and (j + 1)-th level: )(max)(
1
rPrPjj rrrj
CSE 8337 Spring 2009 18
Interpolated precision Idea: If locally precision increases with
increasing recall, then you should get to count that…
So you max of precisions to right of value(Need not be at only standard levels.)
CSE 8337 Spring 2009 19
Precision across queries Recall and Precision are calculated for a
specific query. Generally want a value for many
queries. Calculate average precision recall over
a set of queries. Average precision at recall level r:
Nq – number of queries Pi(r) - precision at recall level r for ith
query
1
( )( )qN
i
i q
P rP rN
CSE 8337 Spring 2009 20
Average Recall/Precision Curve
Typically average performance over a large set of queries.
Compute average precision at each standard recall level across all queries.
Plot average precision/recall curves to evaluate overall system performance on a document/query corpus.
CSE 8337 Spring 2009 21
Compare Two or More Systems The curve closest to the upper right-
hand corner of the graph indicates the best performance
00.2
0.40.6
0.81
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1Recall
Prec
ision
NoStem Stem
CSE 8337 Spring 2009 22
Operating Characteristic Curve (ROC Curve)
CSE 8337 Spring 2009 23
ROC Curve Data False Positive Rate vs True Positive
Rate True Positive Rate
Sensitivity Recall tn/(fp+tn)
False Positive Rate fp/(fp+tn) 1-Specificity Specificity -
CSE 8337 Spring 2009 24
Yet more evaluation measures…
Mean average precision (MAP) Average of the precision value obtained for
the top k documents, each time a relevant doc is retrieved
Avoids interpolation, use of fixed recall levels
MAP for query collection is arithmetic ave. Macro-averaging: each query counts equally
R-precision If have known (though perhaps incomplete)
set of relevant documents of size Rel, then calculate precision of top Rel docs returned
Perfect system could score 1.0.
CSE 8337 Spring 2009 25
Variance For a test collection, it is usual that a
system does crummily on some information needs (e.g., MAP = 0.1) and excellently on others (e.g., MAP = 0.7)
Indeed, it is usually the case that the variance in performance of the same system across queries is much greater than the variance of different systems on the same query.
That is, there are easy information needs and hard ones!
CSE 8337 Spring 2009 2626
Evaluation Graphs are good, but people want summary measures!
Precision at fixed retrieval level Precision-at-k: Precision of top k results Perhaps appropriate for most of web search: all
people want are good matches on the first one or two results pages
But: averages badly and has an arbitrary parameter of k
11-point interpolated average precision The standard measure in the early 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
CSE 8337 Spring 2009 2727
Typical (good) 11 point precisions
SabIR/Cornell 8A1 11pt precision from TREC 8 (1999)
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Recall
Prec
isio
n
CSE 8337 Spring 2009 28
Computing Recall/Precision Points
For a given query, produce the ranked list of retrievals.
Adjusting a threshold on this ranked list produces different sets of retrieved documents, and therefore different recall/precision measures.
Mark each document in the ranked list that is relevant according to the gold standard.
Compute a recall/precision pair for each position in the ranked list that contains a relevant document.
CSE 8337 Spring 2009 29
R=3/6=0.5; P=3/4=0.75
Computing Recall/Precision Points: An Example (modified from [Salton83])
n doc # relevant1 588 x2 589 x3 5764 590 x5 9866 592 x7 9848 9889 57810 98511 10312 59113 772 x14 990
Let total # of relevant docs = 6Check each new recall point:
R=1/6=0.167; P=1/1=1
R=2/6=0.333; P=2/2=1
R=5/6=0.833; p=5/13=0.38
R=4/6=0.667; P=4/6=0.667
Missing one relevant document.
Never reach 100% recall
CSE 8337 Spring 2009 30
F-Measure One measure of performance that takes
into account both recall and precision. Harmonic mean of recall and precision:
Calculated at a specific document in the ranking.
Compared to arithmetic mean, both need to be high for harmonic mean to be high.
Compromise between precision and recall
PRRPPRF 11
22
CSE 8337 Spring 2009 31
A combined measure: F Combined measure that assesses
precision/recall tradeoff is F measure (weighted harmonic mean):
People usually use balanced F1 measure i.e., with = 1 or = ½
RPPR
RP
F
2
2 )1(1)1(1
1
CSE 8337 Spring 2009 32
F1 and other averagesCombined Measures
0
20
40
60
80
100
0 20 40 60 80 100
Precision (Recall fixed at 70%)
Minimum
Maximum
Arithmetic
Geometric
Harmonic
CSE 8337 Spring 2009 33
E Measure (parameterized F Measure)
A variant of F measure that allows weighting emphasis on precision or recall:
Value of controls trade-off: = 1: Equally weight precision and recall
(E=F). > 1: Weight precision more. < 1: Weight recall more.
PRRPPRE
1
2
2
2
2
)1()1(
CSE 8337 Spring 2009 34
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.
collection the in items tnonrelevan of no. totalretrieved items tnonrelevan of no. Fallout
CSE 8337 Spring 2009 35
Fallout Want fallout to be close to 0. In general want to maximize recall
and minimize fallout. Examine fallout-recall graph. More
systems oriented than recall-precision.
CSE 8337 Spring 2009 36
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).
CSE 8337 Spring 2009 37
Utility Subjective measure Cost-Benefit Analysis for retrieved
documents Cr – Benefit of retrieving relevant document Cnr – Cost of retrieving a nonrelevant
document Crn – Cost of not retrieving a relevant
document Nr – Number of relevant documents retrieved Nnr – Number of nonrelevant documents
retrieved Nrn – Number of relevant documents not
retrieved
Nrn))* (Crn Nnr)*((Cnr - Nr) (Cr Utility
CSE 8337 Spring 2009 38
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.
CSE 8337 Spring 2009 39
Experimental Setup for Benchmarking
Analytical performance evaluation is difficult for document retrieval systems because many characteristics such as relevance, distribution of words, etc., are difficult to describe with mathematical precision.
Performance is measured by benchmarking. That is, the retrieval effectiveness of a system is evaluated on a given set of documents, queries, and relevance judgments.
Performance data is valid only for the environment under which the system is evaluated.
CSE 8337 Spring 2009 40
Benchmarks A benchmark collection contains:
A set of standard documents and queries/topics.
A list of relevant documents for each query.
Standard collections for traditional IR:TREC: http://trec.nist.gov/
Standard document collection
Standard queries
Algorithm under test Evaluation
Standard result
Retrieved result
Precision and recall
CSE 8337 Spring 2009 41
Benchmarking The Problems
Performance data is valid only for a particular benchmark.
Building a benchmark corpus is a difficult task.
Benchmark web corpora are just starting to be developed.
Benchmark foreign-language corpora are just starting to be developed.
CSE 8337 Spring 2009 42
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.
CSE 8337 Spring 2009 43
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.
CSE 8337 Spring 2009 44
CSE 8337 Spring 2009 45
Test Collections
CSE 8337 Spring 2009 46
From document collections 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?
CSE 8337 Spring 2009 47
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?
CSE 8337 Spring 2009 48
Kappa measure for inter-judge (dis)agreement Kappa measure
Agreement measure among judges Designed for categorical judgments Corrects for chance agreement
Kappa = [ P(A) – P(E) ] / [ 1 – P(E) ] P(A) – proportion of time judges agree P(E) – what agreement would be by chance Kappa = 0 for chance agreement, 1 for total
agreement.
CSE 8337 Spring 2009 49
Kappa Measure: ExampleNumber of docs Judge 1 Judge 2
300 Relevant Relevant
70 Nonrelevant Nonrelevant
20 Relevant Nonrelevant
10 Nonrelevant relevant
P(A)? P(E)?
CSE 8337 Spring 2009 50
Kappa Example P(A) = 370/400 = 0.925 P(nonrelevant) = (10+20+70+70)/800 = 0.2125 P(relevant) = (10+20+300+300)/800 = 0.7878 P(E) = 0.2125^2 + 0.7878^2 = 0.665 Kappa = (0.925 – 0.665)/(1-0.665) = 0.776
Kappa > 0.8 = good agreement 0.67 < Kappa < 0.8 -> “tentative conclusions”
(Carletta ’96) Depends on purpose of study For >2 judges: average pairwise kappas
CSE 8337 Spring 2009 51
Interjudge Agreement: TREC 3
CSE 8337 Spring 2009 52
Impact of Inter-judge Agreement Impact on absolute performance measure can be
significant (0.32 vs 0.39) Little impact on ranking of different systems or
relative performance Suppose we want to know if algorithm A is better
than algorithm B A standard information retrieval experiment will
give us a reliable answer to this question.
CSE 8337 Spring 2009 53
Critique of pure relevance Relevance vs Marginal Relevance
A document can be redundant even if it is highly relevant
Duplicates The same information from different sources Marginal relevance is a better measure of
utility for the user. Using facts/entities as evaluation units
more directly measures true relevance. But harder to create evaluation set
CSE 8337 Spring 2009 54
Can we avoid human judgment?
No Makes experimental work hard
Especially on a large scale In some very specific settings, can use proxies
E.g.: for approximate vector space retrieval, we can compare the cosine distance closeness of the closest docs to those found by an approximate retrieval algorithm
But once we have test collections, we can reuse them (so long as we don’t overtrain too badly)
CSE 8337 Spring 2009 55
Evaluation at large search engines
Search engines have test collections of queries and hand-ranked results
Recall is difficult to measure on the web Search engines often use precision at top k, e.g., k = 10 . . . or measures that reward you more for getting rank 1
right than for getting rank 10 right. NDCG (Normalized Cumulative Discounted Gain)
Search engines also use non-relevance-based measures. Clickthrough on first result
Not very reliable if you look at a single clickthrough … but pretty reliable in the aggregate.
Studies of user behavior in the lab A/B testing
CSE 8337 Spring 2009 56
A/B testing Purpose: Test a single innovation Prerequisite: You have a large search engine up and
running. Have most users use old system Divert a small proportion of traffic (e.g., 1%) to the
new system that includes the innovation Evaluate with an “automatic” measure like
clickthrough on first result Now we can directly see if the innovation does
improve user happiness. Probably the evaluation methodology that large
search engines trust most In principle less powerful than doing a multivariate
regression analysis, but easier to understand