2013.02.04- slide 1is 240 – spring 2013 prof. ray larson university of california, berkeley school...

2013.02.04- SLIDE 1IS 240 – Spring 2013

Prof. Ray Larson

University of California, Berkeley

School of Information

Principles of Information Retrieval

Lecture 4: Boolean IR System Elements

2013.02.04- SLIDE 2IS 240 – Spring 2013

Review

• Review– Elements of IR Systems

• Collections, Queries• Text processing and Zipf distribution

– Stemmers and Morphological analysis• Inverted file indexes • IR Models - Introduction to the Boolean

Model

2013.02.04- SLIDE 3IS 240 – Spring 2013

Queries

• A query is some expression of a user’s information needs

• Can take many forms– Natural language description of need– Formal query in a query language

• Queries may not be accurate expressions of the information need– Differences between conversation with a

person and formal query expression

2013.02.04- SLIDE 4IS 240 – Spring 2013

Collections of Documents…

• Documents– A document is a representation of some

aggregation of information, treated as a unit.• Collection

– A collection is some physical or logical aggregation of documents

• Let’s take the simplest case, and say we are dealing with a computer file of plain ASCII text, where each line represents the “UNIT” or document.

2013.02.04- SLIDE 5IS 240 – Spring 2013

How to search that collection?

• Manually?– Cat, more

• Scan for strings?– Grep

• Extract individual words to search???– “tokenize” (a unix pipeline)

• tr -sc ’A-Za-z’ ’\012’ < TEXTFILE | sort | uniq –c– See “Unix for Poets” by Ken Church

• Put it in a DBMS and use pattern matching there…– assuming the lines are smaller than the text size limits

for the DBMS

2013.02.04- SLIDE 6IS 240 – Spring 2013

What about VERY big files?

• Scanning becomes a problem• The nature of the problem starts to change

as the scale of the collection increases• A variant of Parkinson’s Law that applies

to databases is:– Data expands to fill the space available to

store it

2013.02.04- SLIDE 7

Document Processing Steps

IS 240 – Spring 2013

2013.02.04- SLIDE 8IS 240 – Spring 2013

Structure of an IR SystemSearchLine Interest profiles

& QueriesDocuments

& data

Rules of the game =Rules for subject indexing +

Thesaurus (which consists of

Lead-InVocabulary

andIndexing

Language

StorageLine

Potentially Relevant

Documents

Comparison/Matching

Store1: Profiles/Search requests

Store2: Documentrepresentations

Indexing (Descriptive and

Subject)

Formulating query in terms of

descriptors

Storage of profiles

Storage of Documents

Information Storage and Retrieval System

Adapted from Soergel, p. 19

2013.02.04- SLIDE 9IS 240 – Spring 2013

Query Processing

• In order to correctly match queries and documents they must go through the same text processing steps as the documents did when they were stored

• In effect, the query is treated like it was a document

• Exceptions (of course) include things like structured query languages that must be parsed to extract the search terms and requested operations from the query– The search terms must still go through the same text

process steps as the document…

2013.02.04- SLIDE 10IS 240 – Spring 2013

Steps in Query processing

• Parsing and analysis of the query text (same as done for the document text)– Morphological Analysis– Statistical Analysis of text

2013.02.04- SLIDE 11IS 240 – Spring 2013

Plotting Word Frequency by Rank

• Say for a text with 100 tokens• Count

– How many tokens occur 1 time (50)– How many tokens occur 2 times (20) …– How many tokens occur 7 times (10) … – How many tokens occur 12 times (1)– How many tokens occur 14 times (1)

• So things that occur the most often share the highest rank (rank 1).

• Things that occur the fewest times have the lowest rank (rank n).

2013.02.04- SLIDE 12IS 240 – Spring 2013

Many similar distributions…

• Words in a text collection• Library book checkout patterns• Bradford’s and Lotka’s laws.• Incoming Web Page Requests (Nielsen)• Outgoing Web Page Requests (Cunha &

Crovella)• Document Size on Web (Cunha &

Crovella)

2013.02.04- SLIDE 13

Zipf Distribution(linear and log scale)

IS 240 – Spring 2013

2013.02.04- SLIDE 14IS 240 – Spring 2013

Resolving Power (van Rijsbergen 79)

The most frequent words are not the most descriptive.

2013.02.04- SLIDE 15IS 240 – Spring 2013

Other Models

• Poisson distribution• 2-Poisson Model• Negative Binomial• Katz K-mixture

– See Church (SIGIR 1995)

2013.02.04- SLIDE 16IS 240 – Spring 2013

Stemming and Morphological Analysis

• Goal: “normalize” similar words• Morphology (“form” of words)

– Inflectional Morphology• E.g,. inflect verb endings and noun number• Never change grammatical class

– dog, dogs– tengo, tienes, tiene, tenemos, tienen

– Derivational Morphology • Derive one word from another, • Often change grammatical class

– build, building; health, healthy

2013.02.04- SLIDE 17IS 240 – Spring 2013

Stemming and Morphological Analysis

• Goal: “normalize” similar words• Morphology (“form” of words)

– Inflectional Morphology• E.g,. inflect verb endings and noun number• Never change grammatical class

– dog, dogs– tengo, tienes, tiene, tenemos, tienen

– Derivational Morphology • Derive one word from another, • Often change grammatical class

– build, building; health, healthy

2013.02.04- SLIDE 18IS 240 – Spring 2013

Simple “S” stemming

• IF a word ends in “ies”, but not “eies” or “aies”– THEN “ies” “y”

• IF a word ends in “es”, but not “aes”, “ees”, or “oes”– THEN “es” “e”

• IF a word ends in “s”, but not “us” or “ss”– THEN “s” NULL

Harman, JASIS Jan. 1991

2013.02.04- SLIDE 20IS 240 – Spring 2013

Too Aggressive Too Timid

organization/organpolicy/police

execute/executivearm/army

european/europecylinder/cylindrical

create/creationsearch/searcher

Errors Generated by Porter Stemmer (Krovetz 93)

2013.02.04- SLIDE 21IS 240 – Spring 2013

Automated Methods

• Stemmers:– Very dumb rules work well (for English)– Porter Stemmer: Iteratively remove suffixes– Improvement: pass results through a lexicon

• Newer stemmers are configurable (Snowball)– Demo…

• Powerful multilingual tools exist for morphological analysis– PCKimmo, Xerox Lexical technology– Require a grammar and dictionary– Use “two-level” automata– Wordnet “morpher”

2013.02.04- SLIDE 22IS 240 – Spring 2013

Wordnet

• Type “wn word” on a machine where wordnet is installed…

• Large exception dictionary:• Demo

aardwolves aardwolf abaci abacus abacuses abacus abbacies abbacy abhenries abhenry abilities ability abkhaz abkhaz abnormalities abnormality aboideaus aboideau aboideaux aboideau aboiteaus aboiteau aboiteaux aboiteau abos abo abscissae abscissa abscissas abscissa absurdities absurdity…

2013.02.04- SLIDE 23IS 240 – Spring 2013

Using NLP

• Strzalkowski (in Reader)

Text NLP represDbasesearch

TAGGERNLP: PARSER TERMS

2013.02.04- SLIDE 24IS 240 – Spring 2013

Using NLP

INPUT SENTENCEThe former Soviet President has been a local hero ever sincea Russian tank invaded Wisconsin.

TAGGED SENTENCEThe/dt former/jj Soviet/jj President/nn has/vbz been/vbn a/dt local/jj hero/nn ever/rb since/in a/dt Russian/jj tank/nn invaded/vbd Wisconsin/np ./per

2013.02.04- SLIDE 25IS 240 – Spring 2013

Using NLP

TAGGED & STEMMED SENTENCEthe/dt former/jj soviet/jj president/nn have/vbz be/vbn a/dt local/jj hero/nn ever/rb since/in a/dt russian/jj tank/nn invade/vbd wisconsin/np ./per

2013.02.04- SLIDE 26IS 240 – Spring 2013

Using NLP

PARSED SENTENCE

[assert

[[perf [have]][[verb[BE]]

[subject [np[n PRESIDENT][t_pos THE]

[adj[FORMER]][adj[SOVIET]]]]

[adv EVER]

[sub_ord[SINCE [[verb[INVADE]]

[subject [np [n TANK][t_pos A]

[adj [RUSSIAN]]]]

[object [np [name [WISCONSIN]]]]]]]]]

2013.02.04- SLIDE 27IS 240 – Spring 2013

Using NLP

EXTRACTED TERMS & WEIGHTS

President 2.623519 soviet 5.416102

President+soviet 11.556747 president+former 14.594883

Hero 7.896426 hero+local 14.314775

Invade 8.435012 tank 6.848128

Tank+invade 17.402237 tank+russian 16.030809

Russian 7.383342 wisconsin 7.785689

2013.02.04- SLIDE 28IS 240 – Spring 2013

Same Sentence, different sys

Enju ParserROOT ROOT ROOT ROOT -1 ROOT been be VBN VB 5been be VBN VB 5 ARG1 President president NNP NNP 3been be VBN VB 5 ARG2 hero hero NN NN 8a a DT DT 6 ARG1 hero hero NN NN 8a a DT DT 11 ARG1 tank tank NN NN 13local local JJ JJ 7 ARG1 hero hero NN NN 8The the DT DT 0 ARG1 President president NNP NNP 3former former JJ JJ 1 ARG1 President president NNP NNP 3Russian russian JJ JJ 12 ARG1 tank tank NN NN 13Soviet soviet NNP NNP 2 MOD President president NNP NNP 3invaded invade VBD VB 14 ARG1 tank tank NN NN 13invaded invade VBD VB 14 ARG2 Wisconsin wisconsin NNP NNP 15has have VBZ VB 4 ARG1 President president NNP NNP 3has have VBZ VB 4 ARG2 been be VBN VB 5since since IN IN 10 MOD been be VBN VB 5since since IN IN 10 ARG1 invaded invade VBD VB 14ever ever RB RB 9 ARG1 since since IN IN 10

2013.02.04- SLIDE 29IS 240 – Spring 2013

Other Considerations

• Church (SIGIR 1995) looked at correlations between forms of words in texts

2013.02.04- SLIDE 30IS 240 – Spring 2013

Assumptions in IR

• Statistical independence of terms• Dependence approximations

2013.02.04- SLIDE 31IS 240 – Spring 2013

Statistical Independence

Two events x and y are statistically independent if the product of their probability of their happening individually equals their probability of happening together.

2013.02.04- SLIDE 32IS 240 – Spring 2013

Statistical Independence and Dependence

• What are examples of things that are statistically independent?

• What are examples of things that are statistically dependent?

2013.02.04- SLIDE 33IS 240 – Spring 2013

• How likely is a red car to drive by given we’ve seen a black one?

• How likely is the word “ambulence” to appear, given that we’ve seen “car accident”?

• Color of cars driving by are independent (although more frequent colors are more likely)

• Words in text are not independent (although again more frequent words are more likely)

Statistical Independence vs. Statistical Dependence

2013.02.04- SLIDE 34IS 240 – Spring 2013

Lexical Associations

• Subjects write first word that comes to mind– doctor/nurse; black/white (Palermo & Jenkins 64)

• Text Corpora yield similar associations• One measure: Mutual Information (Church and Hanks

89)

• If word occurrences were independent, the numerator and denominator would be equal (if measured across a large collection)

2013.02.04- SLIDE 35IS 240 – Spring 2013

Interesting Associations with “Doctor”

(AP Corpus, N=15 million, Church & Hanks 89)

I(x,y) f(x,y) f(x) x f(y) y

11.311.310.79.49.08.98.7

12830861125

1111105110511052751105621

honorarydoctorsdoctorsdoctorsexamineddoctorsdoctor

621442411546213171407

doctordentistsnursestreatingdoctortreatbills

2013.02.04- SLIDE 36IS 240 – Spring 2013

These associations were likely to happen because the non-doctor words shown here are very commonand therefore likely to co-occur with any noun.

Un-Interesting Associations with “Doctor”

I(x,y) f(x,y) f(x) x f(y) y

0.960.950.93

64112

62128469084716

doctorais

7378511051105

withdoctorsdoctors

2013.02.04- SLIDE 37IS 240 – Spring 2013

Query Processing

• Once the text is in a form to match to the indexes then the fun begins– What approach to use?

• Boolean?• Extended Boolean?• Ranked

– Fuzzy sets?– Vector?– Probabilistic?– Language Models? – Neural nets?

• Most of the next few weeks will be looking at these different approaches

2013.02.04- SLIDE 38IS 240 – Spring 2013

Display and formatting

• Have to present the the results to the user• Lots of different options here, mostly

governed by – How the actual document is stored – And whether the full document or just the

metadata about it is presented

2013.02.04- SLIDE 39IS 240 – Spring 2013

Review




Model

2013.02.04- SLIDE 40IS 240 – Spring 2013

What to do with terms…

• Once terms have been extracted from the documents, they need to be stored in some way that lets you get back to documents that those terms came from

• The most common index structure to do this in IR systems is the “Inverted File”

2013.02.04- SLIDE 41IS 240 – Spring 2013

Boolean Implementation: Inverted Files

• We will look at “Vector files” in detail later. But conceptually, an Inverted File is a vector file “inverted” so that rows become columns and columns become rows

2013.02.04- SLIDE 42IS 240 – Spring 2013

How Are Inverted Files Created

• Documents are parsed to extract words (or stems) and these are saved with the Document ID.

Now is the timefor all good men

to come to the aidof their country

Doc 1

It was a dark andstormy night in

the country manor. The time was past midnight

Doc 2

TextProcSteps

2013.02.04- SLIDE 43IS 240 – Spring 2013

How Inverted Files are Created

• After all document have been parsed the inverted file is sorted

2013.02.04- SLIDE 44IS 240 – Spring 2013

How Inverted Files are Created

• Multiple term entries for a single document are merged and frequency information added

2013.02.04- SLIDE 45IS 240 – Spring 2013

Inverted Files• The file is commonly split into a Dictionary

and a Postings file

2013.02.04- SLIDE 46IS 240 – Spring 2013

Inverted Files

• Lots of alternative implementations – E.g.: Cheshire builds within-document

frequency using a hash table during document parsing. Then Document IDs and frequency info are stored in a BerkeleyDB B-tree index keyed by the term.

2013.02.04- SLIDE 47IS 240 – Spring 2013

Btree (conceptual)

B | | D | | F |

AcesBoilers

Cars

F | | P | | Z |

R | | S | | Z |H | | L | | P |

DevilsMinors

PanthersSeminoles

Flyers

HawkeyesHoosiers

2013.02.04- SLIDE 48IS 240 – Spring 2013

Btree with Postings

B | | D | | F |

AcesBoilers

Cars

F | | P | | Z |

R | | S | | Z |H | | L | | P |

DevilsMinors

PanthersSeminoles

FlyersHawkeyesHoosiers

2,4,8,122,4,8,122,4,8,12

2,4,8,12

2,4,8,12

2,4,8,125, 7, 200

2,4,8,122,4,8,128,120

2013.02.04- SLIDE 49IS 240 – Spring 2013

Inverted files

• Permit fast search for individual terms• Search results for each term is a list of

document IDs (and optionally, frequency, part of speech and/or positional information)

• These lists can be used to solve Boolean queries:– country: d1, d2– manor: d2– country and manor: d2

2013.02.04- SLIDE 50IS 240 – Spring 2013

Review




Model

2013.02.04- SLIDE 51IS 240 – Spring 2013

Now we have a system…

• Except for the matching and ranking between the query representation and the document representation– Stored in the inverted files

• We will start to take a look at one model for matching today

• The Boolean Model

2013.02.04- SLIDE 52IS 240 – Spring 2013

IR Models

• Set Theoretic Models– Boolean– Fuzzy– Extended Boolean

• Vector Models (Algebraic)• Probabilistic Models (probabilistic)• Others (e.g., neural networks, etc.)

2013.02.04- SLIDE 53IS 240 – Spring 2013

Boolean Model for IR

• Based on Boolean Logic (Algebra of Sets).• Fundamental principles established by

George Boole in the 1850’s• Deals with set membership and operations

on sets• Set membership in IR systems is usually

based on whether (or not) a document contains a keyword (term)

2013.02.04- SLIDE 54IS 240 – Spring 2013

• Intersection – Boolean ‘AND’ -- -- • Union – Boolean ‘OR’ -- --• Negation – Boolean ‘NOT’ -- --

– Usually means “AND NOT” in IR • Exclusive OR – ‘XOR’ – seldom used,

– Instead

Boolean Operations on Sets

2013.02.04- SLIDE 55IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 56IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 57IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 58IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 59IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 60IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 61IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 62IS 240 – Spring 2013

Boolean Logic

A B

2013.02.04- SLIDE 63IS 240 – Spring 2013

Query Languages

• A way to express the query (formal expression of the information need)

• Types: – Boolean– Natural Language– Stylized Natural Language– Form-Based (GUI)

2013.02.04- SLIDE 64IS 240 – Spring 2013

Simple query language: Boolean

• Terms + Boolean operators– terms

• words• normalized (stemmed) words• phrases• thesaurus terms

– Operators• AND• OR• NOT

– parentheses (for grouping operations)

2013.02.04- SLIDE 65IS 240 – Spring 2013

Boolean Queries

• Cat

• Cat OR Dog

• Cat AND Dog

• (Cat AND Dog)

• (Cat AND Dog) OR Collar

• (Cat AND Dog) OR (Collar AND Leash)

• (Cat OR Dog) AND (Collar OR Leash)

2013.02.04- SLIDE 66IS 240 – Spring 2013

Boolean Queries

• (Cat OR Dog) AND (Collar OR Leash)– Each of the following combinations works:

2013.02.04- SLIDE 67IS 240 – Spring 2013

Boolean Queries

• (Cat OR Dog) AND (Collar OR Leash)– None of the following combinations works:

2013.02.04- SLIDE 68IS 240 – Spring 2013

Boolean Queries

• Usually expressed as INFIX operators in IR– ((a AND b) OR (c AND b))

• NOT is UNARY PREFIX operator– ((a AND b) OR (c AND (NOT b)))

• AND and OR can be n-ary operators– (a AND b AND c AND d)

• Some rules - (De Morgan revisited)– NOT(a) AND NOT(b) = NOT(a OR b)– NOT(a) OR NOT(b)= NOT(a AND b)– NOT(NOT(a)) = a

2013.02.04- SLIDE 69IS 240 – Spring 2013

Boolean Searching

Formal Query:cracks AND beamsAND Width_measurementAND Prestressed_concrete

Cracks

Beams Widthmeasurement

Prestressedconcrete

Relaxed Query:(C AND B AND P) OR(C AND B AND W) OR(C AND W AND P) OR(B AND W AND P)

Relaxed Query:(C AND B AND P) OR(C AND B AND W) OR(C AND W AND P) OR(B AND W AND P)

2013.02.04- SLIDE 70IS 240 – Spring 2013

Boolean Logic

t3

t1 t2

D1D2

D3

D4D5

D6

D8D7

D9

D10

D11

m1

m2

m3m5

m4

m7m8

m6

m2 = t1 t2 t3

m1 = t1 t2 t3

m4 = t1 t2 t3

m3 = t1 t2 t3

m6 = t1 t2 t3

m5 = t1 t2 t3

m8 = t1 t2 t3

m7 = t1 t2 t3

2013.02.04- SLIDE 71IS 240 – Spring 2013

Precedence Ordering

• In what order do we evaluate the components of the Boolean expression?– Parenthesis get done first

• (a or b) and (c or d)• (a or (b and c) or d)

– Usually start from the left and work right (in case of ties)

– Usually (if there are no parentheses)• NOT before AND• AND before OR

2013.02.04- SLIDE 72IS 240 – Spring 2013

Next Time

• More on the Boolean Model including facetted searching, query parse trees and extended Boolean approaches