127
Boolean model and Inverted index Processing Boolean queries Why ranked retrieval? Introduction to Information Retrieval http://informationretrieval.org IIR 1: Boolean Retrieval Hinrich Sch¨ utze Institute for Natural Language Processing, University of Stuttgart 2011-08-29 Sch¨ utze: Boolean retrieval 1 / 30

Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

  • Upload
    others

  • View
    9

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Introduction to Information Retrievalhttp://informationretrieval.org

IIR 1: Boolean Retrieval

Hinrich Schutze

Institute for Natural Language Processing, University of Stuttgart

2011-08-29

Schutze: Boolean retrieval 1 / 30

Page 2: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Models and Methods

1 Boolean model and its limitations (30)

2 Vector space model (30)

3 Probabilistic models (30)

4 Language model-based retrieval (30)

5 Latent semantic indexing (30)

6 Learning to rank (30)

Schutze: Boolean retrieval 3 / 30

Page 3: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Models and Methods

1 Boolean model and its limitations (30)

2 Vector space model (30)

3 Probabilistic models (30)

4 Language model-based retrieval (30)

5 Latent semantic indexing (30)

6 Learning to rank (30)

Schutze: Boolean retrieval 3 / 30

Page 4: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Take-away

Schutze: Boolean retrieval 4 / 30

Page 5: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Take-away

Boolean model and Inverted index: The Boolean model andthe basic data structure of most IR systems

Schutze: Boolean retrieval 4 / 30

Page 6: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Take-away

Boolean model and Inverted index: The Boolean model andthe basic data structure of most IR systems

Processing Boolean queries

Schutze: Boolean retrieval 4 / 30

Page 7: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Take-away

Boolean model and Inverted index: The Boolean model andthe basic data structure of most IR systems

Processing Boolean queries

Why is Boolean retrieval not enough? or Why do we needranked retrieval?

Schutze: Boolean retrieval 4 / 30

Page 8: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Outline

1 Boolean model and Inverted index

2 Processing Boolean queries

3 Why ranked retrieval?

Schutze: Boolean retrieval 5 / 30

Page 9: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 10: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 11: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 12: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 13: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 14: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 15: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

Schutze: Boolean retrieval 6 / 30

Page 16: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Definition of information retrieval

Information retrieval (IR) is finding material (usually documents) ofan unstructured nature (usually text) that satisfies an informationneed from within large collections (usually stored on computers).

The adhoc retrieval problem: Given a user information need and acollection of documents, the IR system determines how well thedocuments satisfy the query and returns a subset of relevantdocuments to the user.

Schutze: Boolean retrieval 6 / 30

Page 17: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean retrieval

Schutze: Boolean retrieval 7 / 30

Page 18: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean retrieval

The Boolean model is arguably the simplest model to base aninformation retrieval system on.

Schutze: Boolean retrieval 7 / 30

Page 19: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean retrieval

The Boolean model is arguably the simplest model to base aninformation retrieval system on.

Queries are Boolean expressions, e.g., Caesar and Brutus

Schutze: Boolean retrieval 7 / 30

Page 20: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean retrieval

The Boolean model is arguably the simplest model to base aninformation retrieval system on.

Queries are Boolean expressions, e.g., Caesar and Brutus

The seach engine returns all documents that satisfy theBoolean expression.

Schutze: Boolean retrieval 7 / 30

Page 21: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Model collection: The works of Shakespeare

Schutze: Boolean retrieval 8 / 30

Page 22: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Model collection: The works of Shakespeare

Schutze: Boolean retrieval 8 / 30

Page 23: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Model collection: The works of Shakespeare

Schutze: Boolean retrieval 8 / 30

Page 24: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Model collection: The works of Shakespeare

Each of Shakespeare’s tragedies,comedies etc is a document in thiscollection.

Schutze: Boolean retrieval 8 / 30

Page 25: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Term-document incidence matrix

Schutze: Boolean retrieval 9 / 30

Page 26: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Term-document incidence matrix

Anthony Julius The Hamlet Othello Macbeth . . .and Caesar Tempest

CleopatraAnthony 1 1 0 0 0 1Brutus 1 1 0 1 0 0Caesar 1 1 0 1 1 1Calpurnia 0 1 0 0 0 0Cleopatra 1 0 0 0 0 0mercy 1 0 1 1 1 1worser 1 0 1 1 1 0. . .Entry is 1 if term occurs. Example: Calpurnia occurs in Julius Caesar.Entry is 0 if term doesn’t occur. Example: Calpurnia doesn’t occur in The

tempest.

Schutze: Boolean retrieval 9 / 30

Page 27: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Term-document incidence matrix

Anthony Julius The Hamlet Othello Macbeth . . .and Caesar Tempest

CleopatraAnthony 1 1 0 0 0 1Brutus 1 1 0 1 0 0Caesar 1 1 0 1 1 1Calpurnia 0 1 0 0 0 0Cleopatra 1 0 0 0 0 0mercy 1 0 1 1 1 1worser 1 0 1 1 1 0. . .Entry is 1 if term occurs. Example: Calpurnia occurs in Julius Caesar.Entry is 0 if term doesn’t occur. Example: Calpurnia doesn’t occur in The

tempest.

Schutze: Boolean retrieval 9 / 30

Page 28: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Term-document incidence matrix

Anthony Julius The Hamlet Othello Macbeth . . .and Caesar Tempest

CleopatraAnthony 1 1 0 0 0 1Brutus 1 1 0 1 0 0Caesar 1 1 0 1 1 1Calpurnia 0 1 0 0 0 0Cleopatra 1 0 0 0 0 0mercy 1 0 1 1 1 1worser 1 0 1 1 1 0. . .Entry is 1 if term occurs. Example: Calpurnia occurs in Julius Caesar.Entry is 0 if term doesn’t occur. Example: Calpurnia doesn’t occur in The

tempest.

Schutze: Boolean retrieval 9 / 30

Page 29: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Term-document incidence matrix

Anthony Julius The Hamlet Othello Macbeth . . .and Caesar Tempest

CleopatraAnthony 1 1 0 0 0 1Brutus 1 1 0 1 0 0Caesar 1 1 0 1 1 1Calpurnia 0 1 0 0 0 0Cleopatra 1 0 0 0 0 0mercy 1 0 1 1 1 1worser 1 0 1 1 1 0. . .Entry is 1 if term occurs. Example: Calpurnia occurs in Julius Caesar.Entry is 0 if term doesn’t occur. Example: Calpurnia doesn’t occur in The

tempest.We will return to this matrix many times in this class.

Schutze: Boolean retrieval 9 / 30

Page 30: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

We can’t build the incidence matrix for large collections

Schutze: Boolean retrieval 10 / 30

Page 31: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

We can’t build the incidence matrix for large collections

Size of incidence matrix: number of documents times numberterms → too large for large collections

Schutze: Boolean retrieval 10 / 30

Page 32: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

We can’t build the incidence matrix for large collections

Size of incidence matrix: number of documents times numberterms → too large for large collections

But the matrix is very sparse – mostly 0s, few 1s.

Schutze: Boolean retrieval 10 / 30

Page 33: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

We can’t build the incidence matrix for large collections

Size of incidence matrix: number of documents times numberterms → too large for large collections

But the matrix is very sparse – mostly 0s, few 1s.

Inverted index: We only record the 1s.

Schutze: Boolean retrieval 10 / 30

Page 34: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Inverted Index

Schutze: Boolean retrieval 11 / 30

Page 35: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Inverted Index

For each term t, we store a list of all documents that contain t.

= For each term t, we store the 1s in its row in the incidencematrix

Brutus −→ 1 2 4 11 31 45 173 174

Caesar −→ 1 2 4 5 6 16 57 132 . . .

Calpurnia −→ 2 31 54 101

...

︸ ︷︷ ︸ ︸ ︷︷ ︸

dictionary postings

Schutze: Boolean retrieval 11 / 30

Page 36: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Outline

1 Boolean model and Inverted index

2 Processing Boolean queries

3 Why ranked retrieval?

Schutze: Boolean retrieval 12 / 30

Page 37: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Schutze: Boolean retrieval 13 / 30

Page 38: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

Schutze: Boolean retrieval 13 / 30

Page 39: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:

Schutze: Boolean retrieval 13 / 30

Page 40: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary

Schutze: Boolean retrieval 13 / 30

Page 41: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary2 Retrieve its postings list from the postings file

Schutze: Boolean retrieval 13 / 30

Page 42: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary2 Retrieve its postings list from the postings file3 Locate Calpurnia in the dictionary

Schutze: Boolean retrieval 13 / 30

Page 43: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary2 Retrieve its postings list from the postings file3 Locate Calpurnia in the dictionary4 Retrieve its postings list from the postings file

Schutze: Boolean retrieval 13 / 30

Page 44: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary2 Retrieve its postings list from the postings file3 Locate Calpurnia in the dictionary4 Retrieve its postings list from the postings file5 Intersect the two postings lists

Schutze: Boolean retrieval 13 / 30

Page 45: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Simple conjunctive query (two terms)

Consider the query: Brutus AND Calpurnia

To find all matching documents using inverted index:1 Locate Brutus in the dictionary2 Retrieve its postings list from the postings file3 Locate Calpurnia in the dictionary4 Retrieve its postings list from the postings file5 Intersect the two postings lists6 Return intersection to user

Schutze: Boolean retrieval 13 / 30

Page 46: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒

Schutze: Boolean retrieval 14 / 30

Page 47: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒

Schutze: Boolean retrieval 14 / 30

Page 48: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒

Schutze: Boolean retrieval 14 / 30

Page 49: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2

Schutze: Boolean retrieval 14 / 30

Page 50: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2

Schutze: Boolean retrieval 14 / 30

Page 51: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2

Schutze: Boolean retrieval 14 / 30

Page 52: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2

Schutze: Boolean retrieval 14 / 30

Page 53: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

Schutze: Boolean retrieval 14 / 30

Page 54: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

Schutze: Boolean retrieval 14 / 30

Page 55: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

Schutze: Boolean retrieval 14 / 30

Page 56: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

Schutze: Boolean retrieval 14 / 30

Page 57: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

Schutze: Boolean retrieval 14 / 30

Page 58: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Intersecting two postings lists

Brutus −→ 1 → 2 → 4 → 11 → 31 → 45 → 173 → 174

Calpurnia −→ 2 → 31 → 54 → 101

Intersection =⇒ 2 → 31

This is linear in the length of the postings lists.

Schutze: Boolean retrieval 14 / 30

Page 59: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

Schutze: Boolean retrieval 15 / 30

Page 60: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Schutze: Boolean retrieval 15 / 30

Page 61: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.

Schutze: Boolean retrieval 15 / 30

Page 62: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.

Schutze: Boolean retrieval 15 / 30

Page 63: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Schutze: Boolean retrieval 15 / 30

Page 64: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Primary commercial retrieval tool for 3 decades

Schutze: Boolean retrieval 15 / 30

Page 65: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Primary commercial retrieval tool for 3 decades

Many professional searchers (e.g., lawyers) still like Booleanqueries.

Schutze: Boolean retrieval 15 / 30

Page 66: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Primary commercial retrieval tool for 3 decades

Many professional searchers (e.g., lawyers) still like Booleanqueries.

You know exactly what you are getting.

Schutze: Boolean retrieval 15 / 30

Page 67: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Primary commercial retrieval tool for 3 decades

Many professional searchers (e.g., lawyers) still like Booleanqueries.

You know exactly what you are getting.

Many search systems you use are also Boolean: search systemon your laptop, in your email reader, on the intranet etc

Schutze: Boolean retrieval 15 / 30

Page 68: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Boolean queries

The example was a simple conjunctive query . . .

. . . the Boolean retrieval model can answer any query that is aBoolean expression.

Boolean queries are queries that use and, or and not to joinquery terms.Views each document as a set of terms.Is precise: Document matches condition or not.

Primary commercial retrieval tool for 3 decades

Many professional searchers (e.g., lawyers) still like Booleanqueries.

You know exactly what you are getting.

Many search systems you use are also Boolean: search systemon your laptop, in your email reader, on the intranet etc

So are we done?

Schutze: Boolean retrieval 15 / 30

Page 69: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Outline

1 Boolean model and Inverted index

2 Processing Boolean queries

3 Why ranked retrieval?

Schutze: Boolean retrieval 16 / 30

Page 70: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Schutze: Boolean retrieval 17 / 30

Page 71: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Schutze: Boolean retrieval 17 / 30

Page 72: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Schutze: Boolean retrieval 17 / 30

Page 73: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Schutze: Boolean retrieval 17 / 30

Page 74: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Not good for the majority of users

Schutze: Boolean retrieval 17 / 30

Page 75: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Not good for the majority of users

Most users are not capable of writing Boolean queries . . .

Schutze: Boolean retrieval 17 / 30

Page 76: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Not good for the majority of users

Most users are not capable of writing Boolean queries . . .

. . . or they are, but they think it’s too much work.

Schutze: Boolean retrieval 17 / 30

Page 77: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Not good for the majority of users

Most users are not capable of writing Boolean queries . . .

. . . or they are, but they think it’s too much work.

Most users don’t want to wade through 1000s of results.

Schutze: Boolean retrieval 17 / 30

Page 78: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

The Boolean model: Pros and Cons

Key property: Documents either match or don’t.

Good for expert users with precise understanding of theirneeds and of the collection.

Also good for applications: Applications can easily consume1000s of results.

Not good for the majority of users

Most users are not capable of writing Boolean queries . . .

. . . or they are, but they think it’s too much work.

Most users don’t want to wade through 1000s of results.

This is particularly true of web search.

Schutze: Boolean retrieval 17 / 30

Page 79: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Schutze: Boolean retrieval 18 / 30

Page 80: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Schutze: Boolean retrieval 18 / 30

Page 81: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Query 1 (boolean conjunction): [standard user dlink 650]

Schutze: Boolean retrieval 18 / 30

Page 82: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Query 1 (boolean conjunction): [standard user dlink 650]

→ 200,000 hits – feast

Schutze: Boolean retrieval 18 / 30

Page 83: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Query 1 (boolean conjunction): [standard user dlink 650]

→ 200,000 hits – feast

Query 2 (boolean conjunction): [standard user dlink 650 nocard found]

Schutze: Boolean retrieval 18 / 30

Page 84: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Query 1 (boolean conjunction): [standard user dlink 650]

→ 200,000 hits – feast

Query 2 (boolean conjunction): [standard user dlink 650 nocard found]

→ 0 hits – famine

Schutze: Boolean retrieval 18 / 30

Page 85: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Problem with Boolean search: Feast or famine

Boolean queries often result in either too few (=0) or toomany (1000s) results.

Query 1 (boolean conjunction): [standard user dlink 650]

→ 200,000 hits – feast

Query 2 (boolean conjunction): [standard user dlink 650 nocard found]

→ 0 hits – famine

In Boolean retrieval, it takes a lot of skill to come up with aquery that produces a manageable number of hits.

Schutze: Boolean retrieval 18 / 30

Page 86: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Feast or famine: No problem in ranked retrieval

Schutze: Boolean retrieval 19 / 30

Page 87: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Feast or famine: No problem in ranked retrieval

With ranking, large result sets are not an issue.

Schutze: Boolean retrieval 19 / 30

Page 88: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Feast or famine: No problem in ranked retrieval

With ranking, large result sets are not an issue.

Just show the top 10 results and the user won’t beoverwhelmed

Schutze: Boolean retrieval 19 / 30

Page 89: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Feast or famine: No problem in ranked retrieval

With ranking, large result sets are not an issue.

Just show the top 10 results and the user won’t beoverwhelmed

Premise: the ranking algorithm works: More relevant resultsare ranked higher than less relevant results.

Schutze: Boolean retrieval 19 / 30

Page 90: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Schutze: Boolean retrieval 20 / 30

Page 91: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Schutze: Boolean retrieval 20 / 30

Page 92: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape them

Schutze: Boolean retrieval 20 / 30

Page 93: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”

Schutze: Boolean retrieval 20 / 30

Page 94: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview them

Schutze: Boolean retrieval 20 / 30

Page 95: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview themEye-track them

Schutze: Boolean retrieval 20 / 30

Page 96: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview themEye-track themTime them

Schutze: Boolean retrieval 20 / 30

Page 97: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview themEye-track themTime themRecord and count their clicks

Schutze: Boolean retrieval 20 / 30

Page 98: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview themEye-track themTime themRecord and count their clicks

The following slides are from Dan Russell’s 2007 JCDL talk

Schutze: Boolean retrieval 20 / 30

Page 99: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Empirical investigation of the effect of ranking

How can we measure how important ranking is?

Observe what searchers do when they are searching in acontrolled setting

Videotape themAsk them to “think aloud”Interview themEye-track themTime themRecord and count their clicks

The following slides are from Dan Russell’s 2007 JCDL talk

Dan Russell was at the “Uber Tech Lead for Search Quality &User Happiness” at Google.

Schutze: Boolean retrieval 20 / 30

Page 100: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent
Page 101: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent
Page 102: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent
Page 103: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent
Page 104: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent
Page 105: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Schutze: Boolean retrieval 26 / 30

Page 106: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Schutze: Boolean retrieval 26 / 30

Page 107: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

Schutze: Boolean retrieval 26 / 30

Page 108: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

There is a very strong bias to click on the top-ranked page.

Schutze: Boolean retrieval 26 / 30

Page 109: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

There is a very strong bias to click on the top-ranked page.

Even if the top-ranked page is not relevant, 30% of users willclick on it.

Schutze: Boolean retrieval 26 / 30

Page 110: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

There is a very strong bias to click on the top-ranked page.

Even if the top-ranked page is not relevant, 30% of users willclick on it.

→ Getting the ranking right is very important.

Schutze: Boolean retrieval 26 / 30

Page 111: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

There is a very strong bias to click on the top-ranked page.

Even if the top-ranked page is not relevant, 30% of users willclick on it.

→ Getting the ranking right is very important.

→ Getting the top-ranked page right is most important.

Schutze: Boolean retrieval 26 / 30

Page 112: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Importance of ranking: Summary

Viewing abstracts: Users are a lot more likely to read theabstracts of the top-ranked pages (1, 2, 3, 4) than theabstracts of the lower ranked pages (7, 8, 9, 10).

Clicking: Distribution is even more skewed for clicking

There is a very strong bias to click on the top-ranked page.

Even if the top-ranked page is not relevant, 30% of users willclick on it.

→ Getting the ranking right is very important.

→ Getting the top-ranked page right is most important.

Schutze: Boolean retrieval 26 / 30

Page 113: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Take-away

Boolean model and Inverted index: The Boolean model andthe basic data structure of most IR systems

Processing Boolean queries

Why is Boolean retrieval not enough? or Why do we needranked retrieval?

Schutze: Boolean retrieval 27 / 30

Page 114: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Resources

Chapter 1 of Introduction to Information Retrieval

Resources at http://informationretrieval.org/essir2011

List of useful information retrieval resourcesShakespeare search engineDaniel Russell’s home page

Schutze: Boolean retrieval 28 / 30

Page 115: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Exercise

Schutze: Boolean retrieval 29 / 30

Page 116: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Exercise

Does Bing/Google use the Boolean model?

Schutze: Boolean retrieval 29 / 30

Page 117: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Exercise

Does Bing/Google use the Boolean model?Does Spotlight use the Boolean model?

Schutze: Boolean retrieval 29 / 30

Page 118: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Schutze: Boolean retrieval 30 / 30

Page 119: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Schutze: Boolean retrieval 30 / 30

Page 120: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

Schutze: Boolean retrieval 30 / 30

Page 121: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor text

Schutze: Boolean retrieval 30 / 30

Page 122: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)

Schutze: Boolean retrieval 30 / 30

Page 123: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)long queries (n large)

Schutze: Boolean retrieval 30 / 30

Page 124: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)long queries (n large)conjunctive boolean query generates very few hits

Schutze: Boolean retrieval 30 / 30

Page 125: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)long queries (n large)conjunctive boolean query generates very few hits

Simple Boolean vs. Ranking of result set

Schutze: Boolean retrieval 30 / 30

Page 126: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)long queries (n large)conjunctive boolean query generates very few hits

Simple Boolean vs. Ranking of result set

Simple Boolean retrieval returns matching documents in noparticular order.

Schutze: Boolean retrieval 30 / 30

Page 127: Introduction to Information Retrieval ` `%%%`# ` ~~~false [0.5cm] IIR 1: Boolean Retrieval · 2011-08-28 · 3 Probabilistic models (30) 4 Language model-based retrieval (30) 5 Latent

Boolean model and Inverted index Processing Boolean queries Why ranked retrieval?

Does web search engines use the Boolean model?

Default interpretation of a query by web search engines: [w1

w2 . . .wn] is w1 AND w2 AND . . . AND wn

Cases where you get hits that do not contain one of the wi :

anchor textpage contains variant of wi (morphology, spelling correction,synonym)long queries (n large)conjunctive boolean query generates very few hits

Simple Boolean vs. Ranking of result set

Simple Boolean retrieval returns matching documents in noparticular order.Google (and most well designed Boolean engines) rank theresult set – they rank good hits (according to some estimatorof relevance) higher than bad hits.

Schutze: Boolean retrieval 30 / 30