0% found this document useful (0 votes)
241 views34 pages

Chapter 4 IR Models

Relevant documents retrieved are {3, 7, 11, 15} 15

Uploaded by

Yohannes Kefale
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
241 views34 pages

Chapter 4 IR Models

Relevant documents retrieved are {3, 7, 11, 15} 15

Uploaded by

Yohannes Kefale
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 34

Introduction to

Information Storage and Retrieval


Chapter Four: IR models

1
IR Models - Basic Concepts
 Word evidence: Bag of words
• IR systems usually adopt index terms to index and retrieve
documents
• Each document is represented by a set of representative
keywords or index terms (called Bag of Words)

• An index term is a word useful for remembering the


document main themes
• Not all terms are equally useful for representing the document
contents:
• less frequent terms allow identifying a narrower set of
documents
• But No ordering information is attached to the Bag of Words
identified from the document collection.

2
IR Models - Basic Concepts
• One central problem regarding IR systems is the
issue of predicting which documents are relevant
and which are not
• Such a decision is usually dependent on a ranking
algorithm which attempts to establish a simple
ordering of the documents retrieved
• Documents appearning at the top of this ordering
are considered to be more likely to be relevant
• Thus ranking algorithms are at the core of IR systems
• The IR models determine the predictions of what is
relevant and what is not, based on the notion of
relevance implemented by the system
3
IR Models - Basic Concepts
• After preprocessing, N distinct terms remain which are
Unique terms that form the VOCABULARY
• Let
– ki be an index term i & dj be a document j
– K = (k1, k2, …, kN) is the set of all index terms
• Each term, i, in a document or query j, is given a real-
valued weight, wij.
– wij is a weight associated with (ki,dj). If wij = 0 , it
indicates that term does not belong to document dj
• The weight wij quantifies the importance of the index term
for describing the document contents
• Vec(dj) = (w1j, w2j, …, wtj) is a term weighted vector
associated with the document dj
4
Mapping Documents & Queries
 Represent both documents and queries as N-dimensional
vectors in a term-document matrix, which shows occurrence
of terms in the document collection or query
 
 E.g. d j  (t1, j , t2, j ,..., t N , j ); qk  (t1,q , t2,q ,..., t N ,q )
 An entry in the matrix corresponds to the “weight” of a
term in the document
– Document collection is mapped to
T1 T2 …. TN term-by-document matrix
D1 w11 w12 … w1N
– View as vector in multidimensional
D2 w21 w22 … w2N space
: : : :
• Nearby vectors are related
: : : :
DM wM1 wM2 … wMN – Normalize for vector length to avoid
the effect of document length
5
Weighting Terms in Vector Sapce
 The importance of the index terms is represented by weights
associated to them
 Problem: to show the importance of the index term for
describing the document/query contents, what weight we can
assign?
 Solution 1: Binary weights: t=1 if present, 0 otherwise
 Similarity: number of terms in common between the
document and the query
 Problem: Not all terms are equally interesting
 E.g. the vs. dog vs. cat
 Solution: Replace binary weights with non-binary weights
 
6
d j  (w1, j , w2, j ,..., wN , j ); qk  (w1,k , w2,k ,..., wN ,k )
The Boolean Model
• Boolean model is a simple model based on set theory
• Boolean model imposes a binary criterion
for deciding relevance
• Terms are either present or absent. Thus,
wij  {0,1}
• sim(q,dj) = 1, if document satisfies the boolean query
0 otherwise
T1 T2 …. TN
D1 w11 w12 … w1N
- Note that, no weights D2 w21 w22 … w2N
assigned in-between 0 and 1, : : : :
just only values 0 or 1 : : : :
DM wM1 wM2 … wMN
7
The Boolean Model: Example
• Generate the relevant documents retrieved by
the Boolean model for the query :
q = k1  (k2  k3)

k2
k1
d7
d2 d6
d4 d5
d3
d1

k3
8
The Boolean Model: Example
• Given the following determine documents retrieved by the
Boolean model based IR system
• Index Terms: K1, …,K8.
• Documents:
1. D1 = {K1, K2, K3, K4, K5}
2. D2 = {K1, K2, K3, K4}
3. D3 = {K2, K4, K6, K8}
4. D4 = {K1, K3, K5, K7}
5. D5 = {K4, K5, K6, K7, K8}
6. D6 = {K1, K2, K3, K4}
• Query: K1 (K2  K3)
• Answer: {D1, D2, D4, D6} ({D1, D2, D3, D6} {D3, D5})
9
= {D1, D2, D6}
The Boolean Model: Further Example
Given the following three documents, Construct Term
– document matrix and find the relevant
documents retrieved by the Boolean model for
given query • Also find the relevant
• D1: “Shipment of gold damaged in a fire” documents for the
queries:
• D2: “Delivery of silver arrived in a silver truck”
• D3: “Shipment of gold arrived in a truck” • (a) “gold delivery”;
• Query: “gold silver truck” • (b) ship gold;
• (c) “silver truck”
Table below shows document –term (ti) matrix

arrive damage deliver fire gold silver ship truck


D1
D2
D3
query
10
The Boolean Model: Further Example
a) gold silver truck: None
b) gold delivery: None
c) ship gold: D1,D3
d) silver truck: D2
arrive damage deliver fire gold silver ship truck
D1 0 1 0 1 1 0 1 0
D2 1 0 1 0 0 1 0 1
D3 1 0 0 0 1 0 1 1
Query a 0 0 0 0 1 1 0 1
Query b 0 0 1 0 1 0 0 0

Query c 0 0 0 0 1 0 1 0
Query d 0 0 0 0 0 1 0 1
11
Exercise
Given the following three documents with the following
contents:
 D1 = “computer information retrieval”
 D2 = “computer retrieval”
 D3 = “information”
 D4 = “computer information”

 What are the relevant documents retrieved for the


queries:
 Q1 = “information  retrieval”
12
 Q2 = “information  ¬computer”
D1 = “computer information retrieval” Q1 = (information  retrieval)
D2 = “computer retrieval” = {D1,D3,D4}  {D1,D2}
D3 = “information” = {D1}
D4 = “computer information”
Q2 = (information  ¬computer)
computer: {D1,D2,D4} = {D1,D3,D4}  {D3}
¬ computer: {D3} = {D3}
information: {D1,D3,D4}
retrieval: {D1,D2}

computer information retrieval


D1 1 1 1
D2 1 0 1
D3 0 1 0
D4 1 1 0
Q1 0 0 0
Q2 0 0 1
13
Exercise: What are the relevant documents retrieved for the
query: ((Caesar OR milton) AND (swift OR shakespeare))
Doc No Term 1 Term 2 Term 3 Term 4
1 Swift
2 Shakespeare
3 Shakespeare Swift
4 Milton
5 Milton Swift
6 Milton Shakespeare
7 Milton Shakespeare Swift
8 Caesar
9 Caesar Swift
10 Caesar Shakespeare
11 Caesar Shakespeare Swift
12 Caesar Milton
13 Caesar Milton Swift
14 Caesar Milton Shakespeare
15 Caesar Milton Shakespeare Swift
14
Q: ((caesar OR milton) AND (swift OR shakespeare))

Doc No Caesar Milton Shakespeare Swift


1 0 0 0 1 Caesar =
{8,9,10,11,12,13,14,15}
2 0 0 1 0
Milton=
3 0 0 1 1 {4,5,6,7,12,13,14,15}
4 0 1 0 0 Shakespeare=
{2,3,6,7,10,11,14,15}
5 0 1 0 1
Swift=
6 0 1 1 0 {1,3,5,7,9,11,13,15}
7 0 1 1 1
8 1 0 0 0 caesar OR milton
{4,5,6,7,8,9,10,11,12,13,14,15}
9 1 0 0 1
10 1 0 1 0 swift OR shakespeare
11 1 0 1 1 {1,2,3,5,6,7,9,10,11,13,14,15}
12 1 1 0 0
((caesar OR milton) AND (swift OR
13 1 1 0 1 shakespeare))
14 1 1 1 0 {5,6,7,9,10,11,13,14,15}
15 15 1 1 1 1
Drawbacks of the Boolean Model
• Retrieval based on binary decision criteria with no notion
of partial matching
• No ranking of the documents is provided (absence of a
grading scale)
• Information need has to be translated into a Boolean
expression which most users find awkward
• The Boolean queries formulated by the users are most often
too simplistic
• As a consequence, the Boolean model frequently returns
either too few or too many documents in response to a
user query

16
Vector-Space Model
• This is the most commonly used strategy for measuring
relevance of documents for a given query. This is
because,
• Use of binary weights is too limiting
• Non-binary weights provide consideration for partial
matches
• The term weights are used to compute a degree of
similarity between a query and each document
• Ranked set of documents provides for better matching
• The idea behind VSM is that
• the meaning of a document is conveyed by the words
used in that document and the weight it carries
17
Vector-Space Model
To find relevant documens for a given query:
• First, Documents and queries are mapped into term
vector space.
• Note that queries are considered as short document
• Short document mean with few words
• Second, In the vector space, queries and documents are
represented as weighted vectors
• There are different weighting technique; the most
widely used one is computing tf*idf for each term
• Third, similarity measurement is used to rank
documents by the closeness of their vectors to the query.
• Documents are ranked by closeness to the query. Closeness
is determined by a similarity score calculation
18
Term-document matrix.
 A collection of n documents and query can be represented
in the vector space model by a term-document matrix.
 An entry in the matrix corresponds to the “weight” of a term in
the document;
 zero means the term has no significance in the document or
it simply doesn’t exist in the document. Otherwise, wij > 0
whenever ki  dj

T1 T2 …. TN
D1 w11 w21 … w1N
D2 w21 w22 … w2N
: : : :
: : : :
DM wM1 wM2 … wMN
19
Computing weights
• How to compute weight for term i in document j (wij ) and
weight for term i in query q (wiq)?
• A good weight must take into account two effects:
– Quantification of intra-document contents (similarity)
• tf factor, the term frequency within a document
– Quantification of inter-documents separation
(dissimilarity)
• idf factor, the inverse document frequency across
documents
– As a result of which most IR systems are using tf*idf
weighting technique:
20
wij = tf(i,j) * idf(i)
Computing weights
• Let:
• N be the total number of documents in the collection
• ni be the number of documents which contain ki
• freq(i,j) total existence frequency of ki within dj
• A normalized tf factor is given by
• f(i,j) = freq(i,j)/max(freq(j))
• where the maximum is computed over all terms which
occur within the document dj
• The idf factor is computed as
• idf(i) = log (N/ni)
• the log is used to make the values of tf and idf
21
comparable. It can also be interpreted as the amount of
information associated with the term ki.
Computing weights
• The best term-weighting schemes use tf*idf weights
which are given by
wij = tf(i,j) * log(N/ni)

• For the query term weights, a suggestion is


wiq = (0.5 + [0.5 * freq(i,q) / max(freq(q)]) * log(N/ni)

• The vector space model with tf*idf weights is a good


ranking strategy with general collections
• The vector space model is usually as good as the known
ranking alternatives. It is also simple and fast to compute.

22
Example: Computing weights
• A collection includes 10,000 documents
• The term A appears 20 times in a particular
document
• The maximum appearance of any term in this
document is 50
• The term A appears in 2,000 of the collection
documents.

• Compute TF*IDF weight?


• f(i,j) = freq(i,j)/max(freq(l,j)) = 20/50 = 0.4
• idf(i) = log(N/ni) = log (10,000/2,000) = log(5) = 2.32
• wij = f(i,j) * log(N/ni) = 0.4 * 2.32 = 0.928
23
Similarity Measurej
dj


• Sim(q,dj) = cos() q
i
 

n
dj q wi , j qi ,k
sim(d j , q)     i 1

 
n n
dj q w 2 2
q
i 1 i, j i 1 i ,k

• Since wij > 0 and wiq > 0, 0 <= sim(q,dj) <=1


• A document is retrieved even if it matches the query
terms only partially
24
Vector-Space Model: Example
• Suppose we query for the query: Q: “gold silver
truck”. The database collection consists of three
documents with the following documents.
• D1: “Shipment of gold damaged in a fire”
• D2: “Delivery of silver arrived in a silver truck”
• D3: “Shipment of gold arrived in a truck”
• Assume that all terms are used, including common
terms, stop words, and also no terms are reduced
to root terms.
• Show retrieval results in ranked order?
25
Vector-Space Model: Example
Terms Q Counts TF DF IDF Wi = TF*IDF
D1 D2 D3 Q D1 D2 D3
a 0 1 1 1 3 0 0 0 0 0
arrived 0 0 1 1 2 0.176 0 0 0.176 0.176
damaged 0 1 0 0 1 0.477 0 0.477 0 0
delivery 0 0 1 0 1 0.477 0 0 0.477 0
fire 0 1 0 0 1 0.477 0 0.477 0 0
gold 1 1 0 1 2 0.176 0.176 0.176 0 0.176
in 0 1 1 1 3 0 0 0 0 0
of 0 1 1 1 3 0 0 0 0 0
silver 1 0 2 0 1 0.477 0.477 0 0.954 0
shipment 0 1 0 1 2 0.176 0 0.176 0 0.176
truck 1 0 1 1 2 0.176 0.176 0 0.176 0.176
Vector-Space Model
Terms Q D1 D2 D3
a 0 0 0 0
arrived 0 0 0.176 0.176
damaged 0 0.477 0 0
delivery 0 0 0.477 0
fire 0 0.477 0 0
gold 0.176 0.176 0 0.176
in 0 0 0 0
of 0 0 0 0
silver 0.477 0 0.954 0
shipment 0 0.176 0 0.176
truck 0.176 0 0.176 0.176
Vector-Space Model: Example
• Compute similarity using cosine Sim(q,d1)
• First, for each document and query, compute all vector
lengths (zero terms ignored)
|d1|= 0.477 2  0.477 2  0.1762  0.1762 = 0.517 = 0.719
|d2|= 0.1762  0.477 2  0.1762  0.1762 = 1.2001 = 1.095
|d3|=
0.1762  0.1762  0.1762  0.1762 = 0.124 = 0.352
|q|= 0.1762  0.4712  0.1762 = 0.2896 = 0.538
• Next, compute dot products (zero products ignored)
Q*d1= 0.176*0.167 = 0.0310
Q*d2 = 0.954*0.477 + 0.176 *0.176 = 0.4862
Q*d3 = 0.176*0.167 + 0.176*0.167 = 0.0620
Vector-Space Model: Example
Now, compute similarity score
Sim(q,d1) = (0.0310) / (0.538*0.719) = 0.0801
Sim(q,d1) = (0.4862 ) / (0.538*1.095)= 0.8246
Sim(q,d1) = (0.0620) / (0.538*0.352)= 0.3271
Finally, we sort and rank documents in descending
order according to the similarity scores
Rank 1: Doc 2 = 0.8246
Rank 2: Doc 3 = 0.3271
Rank 3: Doc 1 = 0.0801

• Exercise: using normalized TF, rank documents using


cosine similarity measure? Hint: Normalize TF of term i
29 in doc j using max frequency of a term k in document j.
Vector-Space Model
• Advantages:
• term-weighting improves quality of the answer set
since it displays in ranked order
• partial matching allows retrieval of documents that
approximate the query conditions
• cosine ranking formula sorts documents according
to degree of similarity to the query

• Disadvantages:
• assumes independence of index terms (??)

30
More Example
Suppose the database collection consists of the following documents.
c1: Human machine interface for Lab ABC computer
applications
c2: A survey of user opinion of computer system response time
c3: The EPS user interface management system
c4: System and human system engineering testing of EPS
c5: Relation of user-perceived response time to error measure
M1: The generation of random, binary, unordered trees
M2: The intersection graph of paths in trees
M3: Graph minors: Widths of trees and well-quasi-ordering
M4: Graph minors: A survey
Query:
Find documents relevant to "human computer interaction”
31
Exercises
 Given the following documents, rank documents according to
their relevance to the query using Cosine similarity, Euclidean
distance and Inner product measures?

docID words in document


1 Taipei Taiwan
2 Macao Taiwan Shanghai
3 Japan Sapporo
4 Sapporo Osaka Taiwan

Query: Taiwan Taiwan Sapporo ?

32
End of Chapter 4

33
Test
• Given the following Term-Document matrix and Query, perform:
1. Eucledian Distance between the Query and Each Document
2. Inner Product between the Query and Each Document
economy develop country
D1 1 3 2
D2 3 2 1
D3 2 1 0
Q 1 1 0

34

You might also like