Acknowledgements to Pandu Nayak and
Prabhakar Raghavan of Stanford, Hinrich
Schütze and Christina Lioma of Stutgart, Lee
Giles (and his sources) of Penn State
• You have learned to crawl the web
– Being a good citizen and not hurting the servers
– Extracting the kind of information that you want
– Storing the retrieved material locally
• Now, what are you going to do with those materials?
– Develop a way to retrieve what you want from that collection, as you need it.
• Information Retrieval:
– Given a collection of “documents”
– Retrieve
• One or more documents that contain the specific information you want
• Obtain the answer to an information need by querying the document collection
• Introduce the concepts of Information
Retrieval
– Assume you have a collection
– How to look into the documents for your information need
– How to do it efficiently
• Tools to build indices
– Apache Lucene
– Apache Solr
• Recall our primary resource:
– Christopher D. Manning, Prabhakar Raghavan and
Hinrich Schütze, Introduction to Information Retrieval,
Cambridge University Press. 2008.
• The entire book is available online, free, at http://nlp.stanford.edu/IR-book/informationretrieval-book.html
• I will use some of the slides that they provide to go with the book.
• I will also use slides from other sources and make new ones. The primary other source is Dr. Lee
Giles, Penn State. Information Retrieval IST441
• Information Retrieval (IR) is finding material
(usually documents) of an unstructured nature (usually text) that satisfies an information need from within large collections (usually stored on computers).
• Note the use of the word “usually.” We will see examples where the material is not documents, and not text.
• IR is about finding a needle in a haystack
– finding some particular thing in a very large collection of similar things.
• Our examples are necessarily small, so that we can comprehend them. Do remember, that all that we say must scale to very large quantities.
• Which plays of Shakespeare contain the words
Brutus AND Caesar but NOT Calpurnia?
– See http://www.rhymezone.com/shakespeare/
• One could grep all of Shakespeare’s plays for
Brutus and Caesar, then strip out lines containing Calpurnia?
• Why is that not the answer?
– Slow (for large corpora)
– NOT Calpurnia is non-trivial
– Other operations (e.g., find the word Romans near
countrymen) not feasible
– Ranked retrieval (best documents to return)
• Go to http://www.rhymezone.com/shakespeare/
• Enter, one at a time,
– Caesar
– Brutus
– Calpurnia
• For each, note the collection of plays that contain the term
• We need to associate one or more plays with each term – Term Incidence
For our example, we will use a subset of the plays
First approach – make a matrix with terms on one axis and plays on the other
Brutus AND Caesar BUT NOT
Calpurnia
Antony
Brutus
Caesar
Calpurnia
Cleopatra mercy worser
All the plays
Antony and Cleopatra Julius Caesar The Tempest Hamlet Othello Macbeth
1
1
1
0
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
1
1
0
1
1
0
0
1
1
0
0
1
0
0
1
1
1
0
1
0
0
1
0
1 if play contains word , 0 otherwise
• So we have a 0/1 vector for each term.
• To answer query: take the vectors for Brutus,
Caesar and Calpurnia (complemented)
bitwise AND.
• 110100 AND 110111 AND 101111 =
100100.
Reminder: 0 AND 0 = 0, 0 AND 1 =0, 1 AND 0 = 0, 1 AND 1 = 1
The effect of the AND operation is 1 if the items are both 1 and 0 if they are not both 1
Agrippa [Aside to DOMITIUS ENOBARBUS]: Why,
Enobarbus,
When Antony found Julius Caesar dead,
He cried almost to roaring; and he wept
When at Philippi he found Brutus slain.
Lord Polonius: I did enact Julius Caesar I was killed i' the Capitol;
Brutus killed me.
• Try another one
• What is the vector for the query
– Antony and mercy
• What would we do to find Antony OR mercy?
• Collection : Fixed set of documents
• Goal : Retrieve documents with information that is relevant to the user’s information need and helps the user complete a task
Ultimately, some task to perform.
TASK
Info
Need
Verbal form
Some information is required in order to perform the task.
The information need must be expressed in words (usually).
Query
It may be necessary to rephrase the query and try again
Query
Refinement
The information need must be expressed in the form of a query that can be processed.
SEARCH
ENGINE
Results
Corpus
TASK
Info
Need
Potential pitfalls between task and query results
Get rid of mice in a politically correct way
Misconception?
Info about removing mice without killing them
Mistranslation?
Verbal form
How do I trap mice alive?
Misformulation?
mouse trap Query
SEARCH
ENGINE
Query
Refinement
Results
Corpus
• Precision: How well do the results match the information need?
• Recall: What fraction of the available correct results were retrieved?
• These are the basic concepts of information retrieval evaluation.
• Consider N = 1 million documents, each with about 1000 words.
• Avg 6 bytes/word including spaces/punctuation
– 6GB of data in the documents.
• Say there are M = 500K distinct terms among these.
• 500K x 1M matrix has half-atrillion 0’s and 1’s.
• But it has no more than one billion
1’s.
– matrix is extremely sparse.
• What’s a better representation?
– We only record the 1 positions.
– i.e. We don’t need to know which documents do not have a term, only those that do.
Why?
• For each term t, we must store a list of all documents that contain t.
– Identify each by a docID, a document serial number
• Can we used fixed-size arrays for this?
Brutus 1 2 4 11 31 45 173 174
Caesar 1 2 4 5 6 16 57 132
Calpurnia
2 31 54 101
What happens if the word Caesar is added to document 14?
More likely, add document 14, which contains “Caesar.”
We need variable-size postings lists
On disk, a continuous run of postings is normal and best
In memory, can use linked lists or variable length arrays
Some tradeoffs in size/ease of insertion
Posting
Brutus
Caesar
Calpurnia
Dictionary
2
1 2
1
4 11 31
31
2 4 5 6
54 101
45 173
16
174
57 132
Postings
22 Sorted by docID (more later on why).
Sec. 1.2
Documents to be indexed.
Friends, Romans, countrymen.
Token stream.
Tokenizer
Friends Romans
Modified tokens.
Stop words, stemming, capitalization, cases, etc.
Inverted index.
Linguistic modules
Indexer friend friend roman countryman roman
Countrymen
2
1
13 countryman
4
2
16
Sequence of (Modified token, Document
ID) pairs .
Initially, all the tokens from document 1, then all the tokens from document 2, etc., without regard for duplication.
Doc 1 Doc 2
I did enact Julius
Caesar I was killed i' the Capitol;
Brutus killed me.
So let it be with
Caesar. The noble
Brutus hath told you
Caesar was ambitious
docID within terms
Core indexing step
Multiple term entries in a single document are merged.
Split into
Dictionary and Postings
Doc. frequency information is added.
Number of documents in which the term appears
Lists of docIDs
Terms and counts
Pointers 27
• A small diversion
• Computer storage
– Processor caches
– Main memory
– External storage (hard disk, other devices)
• Very substantial differences in access speeds
– Processor caches mostly used by the operating system for rapid access to data that will be needed soon
– Main memory.
• Limited quantities. High speed access
– Hard disk
• Much larger quantities, speed restricted, access in fixed units (blocks)
• From the iMac
– Memory
• 4GB (two 2GB SO-DIMMs) of 1333MHz DDR3
SDRAM; four SO-DIMM slots support up to
16GB
– Hard drive
• 500GB or 1TB 7200-rpm Serial ATA hard drive
• Optional 2TB 7200-rpm Serial ATA hard drive
These are big numbers, but the potential size of a significant collection is larger still. The steps taken to optimize use of storage are critical to satisfactory response time.
virtual memory (on disk) page
1 page
2 page
3 page
4
... ... ... ... ...
real memory (RAM) slot
1 slot
2 slot
3 slot
4 slot
5 slot
6 slot
7 slot
8 slot
9 slot
10 slot
11 slot
12 virtual page 2 generates a “page fault” when referencing virtual page 71 virtual page 71 is brought from disk into real memory
... ... ... ... ... ... ... ... ...
... ... ... ... ... ... ... ... ...
reference to page 71 b
... ... ... ... ... ... ... ... ...
... ... ... ... ... ...
page
70 page
71 page
72
• Using the index we just built, examine the terms in some order, looking for the terms in the query.
• Consider processing the query:
Brutus AND Caesar
– Locate Brutus in the Dictionary;
• Retrieve its postings.
– Locate Caesar in the Dictionary;
• Retrieve its postings.
– “Merge” the two postings:
2 4 8 16 32 64 128
1 2 3 5 8 13 21 34
Brutus
Caesar
32
2 8
Sec. 1.3
• Walk through the two postings simultaneously, in time linear in the total number of postings entries
8 13
Brutus
Caesar
If the list lengths are x and y, the merge takes O(x+y) operations.
What does that mean?
Crucial: postings sorted by docID .
33
34
• Let’s assume that
– the term mercy appears in documents 1, 2,
13, 18, 24,35, 54
– the term noble appears in documents 1, 5,
7, 13, 22, 24, 56
• Show the document lists, then step through the merge algorithm to obtain the search results.
Sec. 1.3
• The Boolean retrieval model is being able to ask a query that is a Boolean expression:
– Boolean Queries are queries using AND, OR and
NOT to join query terms
• Views each document as a set of words
• Is precise: document matches condition or not.
– Perhaps the simplest model to build
• Primary commercial retrieval tool for 3 decades.
• Many search systems you still use are Boolean:
– Email, library catalog, Mac OS X Spotlight
36
37
• Consider a query that is an and of n terms, n > 2
• For each of the terms, get its postings list, then and them together
• Example query: BRUTUS AND CALPURNIA AND
CAESAR
• What is the best order for processing this query?
37
38
• Example query: BRUTUS AND CALPURNIA AND
CAESAR
• Simple and effective optimization: Process in order of increasing frequency
• Start with the shortest postings list, then keep cutting further
• In this example, first CAESAR , then CALPURNIA , then BRUTUS
38
39
39
40
• Example query: ( MADDING OR CROWD ) and ( IGNOBLE OR STRIFE )
• Get frequencies for all terms
• Estimate the size of each or by the sum of its frequencies (conservative)
• Process in increasing order of or sizes
40
• These basic techniques are pretty simple
• There are challenges
– Scaling
• as everything becomes digitized, how well do the processes scale?
– Intelligent information extraction
• I want information, not just a link to a place that might have that information.
Yotta
• Soon most everything will be recorded and indexed
• Most bytes will never be seen by humans.
• Data summarization, trend detection anomaly detection are key technologies
Gray - Microsoft
See Mike Lesk:
How much information is
there: http://www.lesk.com/mlesk/ksg97/ksg.ht
ml
See Lyman & Varian:
How much information http://www.sims.berkeley.edu/research/projects/howmuch-info /
Everything
Recorded !
All Books
MultiMedia
All books
(words)
A Photo
A movie
Zetta
Exa
Peta
Tera
Giga
Mega
24 Yecto, 21 zepto, 18 atto, 15 femto, 12 pico, 9 nano, 6 micro, 3 milli
Slide source: Lee Giles (modified)
A Book
Kilo
• Dependent on
– Acquiring information
– Storing information
– Indexing
– Interaction with the information source
– Evaluation
• We have learned about
– Acquiring information
– Storing information
– Indexing (the basics)
– Interaction with the information source
– Evaluation
• Still to come
– Acquiring information
– Storing information
– Indexing (more)
– Interaction with the information source
– Evaluation
Unusual and diverse documents
Unusual and diverse users, queries, information needs
Beyond terms, exploit ideas from social networks
link analysis, clickstreams …
How do search engines work? And how can we make them better?
46
• Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze,
Introduction to Information Retrieval, Cambridge University Press. 2008.
– Book available online at http://nlp.stanford.edu/IR-book/information-retrieval-book.html
– Many of these slides are taken directly from the authors’ slides from the first chapter of the book.
• C. Lee Giles, Penn State. http://clgiles.ist.psu.edu/
• Paging figure from Vittore Carsarosa, University of Parma, Italy