Introduction to Information Retrieval
Introduction to
Information Retrieval
CS276
Information Retrieval and Web Search
Pandu Nayak and Prabhakar Raghavan
Lecture 9: Query expansion
Introduction to Information Retrieval
Reminder
Midterm in class on Thursday 28th
Material from first 8 lectures
Open book, open notes
You can use (and should bring!) a basic calculator
You cannot use any wired or wireless
communication. Use of such communication will be
regarded as an Honor Code violation.
You can preload the pdf of the book on to your
laptop which you can use disconnected in the room.
Introduction to Information Retrieval
Recap of the last lecture
Evaluating a search engine
Benchmarks
Precision and recall
Results summaries
Introduction to Information Retrieval
Recap: 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)
Relevant Nonrelevant
Retrieved tp fp
Not Retrieved fn tn
Precision P = tp/(tp + fp)
Recall R = tp/(tp + fn)
4
Introduction to Information Retrieval
Recap: A combined measure: F
Combined measure that assesses precision/recall
tradeoff is F measure (weighted harmonic mean):
( 1) PR
1 2
F
1
(1 )
1 PR
2
P R
People usually use balanced F1 measure
i.e., with = 1 or = ½
Harmonic mean is a conservative average
See CJ van Rijsbergen, Information Retrieval
5
Introduction to Information Retrieval
This lecture
Improving results
For high recall. E.g., searching for aircraft doesn’t match
with plane; nor thermodynamic with heat
Options for improving results…
Global methods
Query expansion
Thesauri
Automatic thesaurus generation
Local methods
Relevance feedback
Pseudo relevance feedback
Introduction to Information Retrieval Sec. 9.1
Relevance Feedback
Relevance feedback: user feedback on relevance of
docs in initial set of results
User issues a (short, simple) query
The user marks some results as relevant or non-relevant.
The system computes a better representation of the
information need based on feedback.
Relevance feedback can go through one or more
iterations.
Idea: it may be difficult to formulate a good query
when you don’t know the collection well, so iterate
Introduction to Information Retrieval Sec. 9.1
Relevance feedback
We will use ad hoc retrieval to refer to regular
retrieval without relevance feedback.
We now look at four examples of relevance feedback
that highlight different aspects.
Introduction to Information Retrieval
Similar pages
Introduction to Information Retrieval Sec. 9.1.1
Relevance Feedback: Example
Image search engine
http://nayana.ece.ucsb.edu/imsearch/imsearch.html
Introduction to Information Retrieval Sec. 9.1.1
Results for Initial Query
Introduction to Information Retrieval Sec. 9.1.1
Relevance Feedback
Introduction to Information Retrieval Sec. 9.1.1
Results after Relevance Feedback
Introduction to Information Retrieval Sec. 9.1.1
Initial query/results
Initial query: New space satellite applications
+ 1. 0.539, 08/13/91, NASA Hasn’t Scrapped Imaging Spectrometer
+ 2. 0.533, 07/09/91, NASA Scratches Environment Gear From Satellite Plan
3. 0.528, 04/04/90, Science Panel Backs NASA Satellite Plan, But Urges Launches of Smaller
Probes
4. 0.526, 09/09/91, A NASA Satellite Project Accomplishes Incredible Feat: Staying Within
Budget
5. 0.525, 07/24/90, Scientist Who Exposed Global Warming Proposes Satellites for Climate
Research
6. 0.524, 08/22/90, Report Provides Support for the Critics Of Using Big Satellites to Study
Climate
7. 0.516, 04/13/87, Arianespace Receives Satellite Launch Pact From Telesat Canada
8. 0.509, 12/02/87, Telecommunications Tale of Two Companies
+
User then marks relevant documents with “+”.
Introduction to Information Retrieval Sec. 9.1.1
Expanded query after relevance feedback
2.074 new 15.106 space
30.816 satellite 5.660 application
5.991 nasa 5.196 eos
4.196 launch 3.972 aster
3.516 instrument 3.446 arianespace
3.004 bundespost 2.806 ss
2.790 rocket 2.053 scientist
2.003 broadcast 1.172 earth
0.836 oil 0.646 measure
Introduction to Information Retrieval Sec. 9.1.1
Results for expanded query
2 1. 0.513, 07/09/91, NASA Scratches Environment Gear From Satellite Plan
1 2. 0.500, 08/13/91, NASA Hasn’t Scrapped Imaging Spectrometer
3. 0.493, 08/07/89, When the Pentagon Launches a Secret Satellite, Space Sleuths Do
Some Spy Work of Their Own
4. 0.493, 07/31/89, NASA Uses ‘Warm’ Superconductors For Fast Circuit
8 5. 0.492, 12/02/87, Telecommunications Tale of Two Companies
6. 0.491, 07/09/91, Soviets May Adapt Parts of SS-20 Missile For Commercial Use
7. 0.490, 07/12/88, Gaping Gap: Pentagon Lags in Race To Match the Soviets In Rocket
Launchers
8. 0.490, 06/14/90, Rescue of Satellite By Space Agency To Cost $90 Million
Introduction to Information Retrieval Sec. 9.1.1
Key concept: Centroid
The centroid is the center of mass of a set of points
Recall that we represent documents as points in a
high-dimensional space
Definition: Centroid
1
(C )
| C | dC
d
where C is a set of documents.
Introduction to Information Retrieval Sec. 9.1.1
Rocchio Algorithm
The Rocchio algorithm uses the vector space model
to pick a relevance feedback query
Rocchio seeks the query qopt that maximizes
qopt arg max [cos( q, (Cr )) cos( q , (Cnr ))]
q
Tries to separate docs marked relevant and non-
relevant 1 1
qopt
C d j
C d j
r d j Cr nr d j Cr
Problem: we don’t know the truly relevant docs
Introduction to Information Retrieval Sec. 9.1.1
The Theoretically Best Query
x x
x x
o x x
x x x x
o x x
x
o x
o x x
o o x
x
x non-relevant documents
Optimal
query o relevant documents
Introduction to Information Retrieval Sec. 9.1.1
Rocchio 1971 Algorithm (SMART)
Used in practice:
1 1
q m q0
Dr d j
Dnr d j
d j Dr d j Dnr
Dr = set of known relevant doc vectors
Dnr = set of known irrelevant doc vectors
Different from Cr and Cnr !
qm = modified query vector; q0 = original query vector; α,β,γ:
weights (hand-chosen or set empirically)
New query moves toward relevant documents and away
from irrelevant documents
Introduction to Information Retrieval Sec. 9.1.1
Subtleties to note
Tradeoff α vs. β/γ : If we have a lot of judged
documents, we want a higher β/γ.
Some weights in query vector can go negative
Negative term weights are ignored (set to 0)
Introduction to Information Retrieval Sec. 9.1.1
Relevance feedback on initial query
Initial
x x
query x
o x
x x
x x
o x x
x
x o
x o x
o o x
x x
x
x known non-relevant documents
Revised
query o known relevant documents
Introduction to Information Retrieval Sec. 9.1.1
Relevance Feedback in vector spaces
We can modify the query based on relevance
feedback and apply standard vector space model.
Use only the docs that were marked.
Relevance feedback can improve recall and
precision
Relevance feedback is most useful for increasing
recall in situations where recall is important
Users can be expected to review results and to take time
to iterate
Introduction to Information Retrieval Sec. 9.1.1
Positive vs Negative Feedback
Positive feedback is more valuable than negative
feedback (so, set < ; e.g. = 0.25, = 0.75).
Many systems only allow positive feedback (=0).
W
hy
?
Introduction to Information Retrieval
Example – RR Computation
q0: cheap CDs cheap DVDs extremely cheap CD
D1: CDs cheap software cheap CDs
D2: cheap thrills DVDs
Assume alpha= 1, beta= 0.75, gamma= 0.25.
25
Introduction to Information Retrieval
Aside: Vector Space can be
Counterintuitive.
Doc
x x
“J. Snow x x
x x
& Cholera” x
o x x x
x x x
q1
x x
x x x x
x
x
x
q1 query “cholera”
Query o www.ph.ucla.edu/epi/snow.html
x other documents
“cholera”
Introduction to Information Retrieval
High-dimensional Vector Spaces
The queries “cholera” and “john snow” are far from
each other in vector space.
How can the document “John Snow and Cholera” be
close to both of them?
Our intuitions for 2- and 3-dimensional space don't
work in >10,000 dimensions.
3 dimensions: If a document is close to many
queries, then some of these queries must be close
to each other.
Doesn't hold for a high-dimensional space.
Introduction to Information Retrieval Sec. 9.1.3
Relevance Feedback: Assumptions
A1: User has sufficient knowledge for initial query.
A2: Relevance prototypes are “well-behaved”.
Term distribution in relevant documents will be similar
Term distribution in non-relevant documents will be
different from those in relevant documents
Either: All relevant documents are tightly clustered around a
single prototype.
Or: There are different prototypes, but they have significant
vocabulary overlap.
Similarities between relevant and irrelevant documents are small
Introduction to Information Retrieval Sec. 9.1.3
Violation of A1
User does not have sufficient initial knowledge.
Examples:
Misspellings (Brittany Speers).
Cross-language information retrieval (hígado).
Mismatch of searcher’s vocabulary vs. collection
vocabulary
Cosmonaut/astronaut
Introduction to Information Retrieval Sec. 9.1.3
Violation of A2
There are several relevance prototypes.
Examples:
Burma/Myanmar
Contradictory government policies
Pop stars that worked at Burger King
Often: instances of a general concept
Good editorial content can address problem
Report on contradictory government policies
Introduction to Information Retrieval
Relevance Feedback: Problems
Long queries are inefficient for typical IR engine.
Long response times for user.
W
High cost for retrieval system. hy
?
Partial solution:
Only reweight certain prominent terms
Perhaps top 20 by term frequency
Users are often reluctant to provide explicit
feedback
It’s often harder to understand why a particular
document was retrieved after applying relevance
feedback
Introduction to Information Retrieval Sec. 9.1.5
Evaluation of relevance feedback
strategies
Use q0 and compute precision and recall graph
Use qm and compute precision recall graph
Assess on all documents in the collection
Spectacular improvements, but … it’s cheating!
Partly due to known relevant documents ranked higher
Must evaluate with respect to documents not seen by user
Use documents in residual collection (set of documents minus those
assessed relevant)
Measures usually then lower than for original query
But a more realistic evaluation
Relative performance can be validly compared
Empirically, one round of relevance feedback is often very useful.
Two rounds is sometimes marginally useful.
Introduction to Information Retrieval Sec. 9.1.5
Evaluation of relevance feedback
Second method – assess only the docs not rated by
the user in the first round
Could make relevance feedback look worse than it really is
Can still assess relative performance of algorithms
Most satisfactory – use two collections each with
their own relevance assessments
q0 and user feedback from first collection
qm run on second collection and measured
Introduction to Information Retrieval Sec. 9.1.3
Evaluation: Caveat
True evaluation of usefulness must compare to other
methods taking the same amount of time.
Alternative to relevance feedback: User revises and
resubmits query.
Users may prefer revision/resubmission to having to
judge relevance of documents.
There is no clear evidence that relevance feedback is
the “best use” of the user’s time.
Introduction to Information Retrieval Sec. 9.1.4
Relevance Feedback on the Web
Some search engines offer a similar/related pages feature (this is a
trivial form of relevance feedback)
Google (link-based)
Altavista
α/β/γ ??
Stanford WebBase
But some don’t because it’s hard to explain to average user:
Alltheweb
bing
Yahoo
Excite initially had true relevance feedback, but abandoned it due
to lack of use.
Introduction to Information Retrieval Sec. 9.1.4
Excite Relevance Feedback
Spink et al. 2000
Only about 4% of query sessions from a user used
relevance feedback option
Expressed as “More like this” link next to each result
But about 70% of users only looked at first page of
results and didn’t pursue things further
So 4% is about 1/8 of people extending search
Relevance feedback improved results about 2/3 of
the time
Introduction to Information Retrieval Sec. 9.1.6
Pseudo relevance feedback
Pseudo-relevance feedback automates the “manual”
part of true relevance feedback.
Pseudo-relevance algorithm:
Retrieve a ranked list of hits for the user’s query
Assume that the top k documents are relevant.
Do relevance feedback (e.g., Rocchio)
Works very well on average
But can go horribly wrong for some queries.
Several iterations can cause query drift.
Why?
Introduction to Information Retrieval Sec. 9.2.2
Query Expansion
In relevance feedback, users give additional input
(relevant/non-relevant) on documents, which is
used to reweight terms in the documents
In query expansion, users give additional input
(good/bad search term) on words or phrases
Introduction to Information Retrieval
Query assist
Would you expect such a feature to increase the query
volume at a search engine?
Introduction to Information Retrieval Sec. 9.2.2
How do we augment the user query?
Manual thesaurus
E.g. MedLine: physician, syn: doc, doctor, MD, medico
Can be query rather than just synonyms
Global Analysis: (static; of all documents in collection)
Automatically derived thesaurus
(co-occurrence statistics)
Refinements based on query log mining
Common on the web
Local Analysis: (dynamic)
Analysis of documents in result set
Introduction to Information Retrieval Sec. 9.2.2
Example of manual thesaurus
Introduction to Information Retrieval Sec. 9.2.2
Thesaurus-based query expansion
For each term, t, in a query, expand the query with synonyms and
related words of t from the thesaurus
feline → feline cat
May weight added terms less than original query terms.
Generally increases recall
Widely used in many science/engineering fields
May significantly decrease precision, particularly with ambiguous
terms.
“interest rate” “interest rate fascinate evaluate”
There is a high cost of manually producing a thesaurus
And for updating it for scientific changes
Introduction to Information Retrieval Sec. 9.2.3
Automatic Thesaurus Generation
Attempt to generate a thesaurus automatically by
analyzing the collection of documents
Fundamental notion: similarity between two words
Definition 1: Two words are similar if they co-occur
with similar words.
Definition 2: Two words are similar if they occur in a
given grammatical relation with the same words.
You can harvest, peel, eat, prepare, etc. apples and
pears, so apples and pears must be similar.
Co-occurrence based is more robust, grammatical
relations are more accurate. Why?
Introduction to Information Retrieval Sec. 9.2.3
Co-occurrence Thesaurus
Simplest way to compute one is based on term-term similarities
in C = AAT where A is term-document matrix.
wi,j = (normalized) weight for (ti ,dj)
dj N
ti What does C
contain if A is
a term-doc
incidence
M (0/1) matrix?
For each ti, pick terms with high values in C
Introduction to Information Retrieval Sec. 9.2.3
Automatic Thesaurus Generation
Example
Introduction to Information Retrieval Sec. 9.2.3
Automatic Thesaurus Generation
Discussion
Quality of associations is usually a problem.
Term ambiguity may introduce irrelevant statistically
correlated terms.
“Apple computer” “Apple red fruit computer”
Problems:
False positives: Words deemed similar that are not
False negatives: Words deemed dissimilar that are similar
Since terms are highly correlated anyway, expansion
may not retrieve many additional documents.
Introduction to Information Retrieval
Indirect relevance feedback
On the web, DirectHit introduced a form of indirect
relevance feedback.
DirectHit ranked documents higher that users look at
more often.
Clicked on links are assumed likely to be relevant
Assuming the displayed summaries are good, etc.
Globally: Not necessarily user or query specific.
This is the general area of clickstream mining
Today – handled as part of machine-learned ranking
Introduction to Information Retrieval
Resources
IIR Ch 9
MG Ch. 4.7
MIR Ch. 5.2 – 5.4
