1 CS 430 / INFO 430 Information Retrieval Lecture 26 Thesauruses and Cluster Analysis 2
Dec 24, 2015
1
CS 430 / INFO 430Information Retrieval
Lecture 26
Thesauruses and Cluster Analysis 2
2
Course Administration
CS 490 and CS 790 Independent Research Projects
• Web Research Infrastructure -- Build a system to bring complete crawls of the Web from the Internet Archive to the Cornell Theory Center and make them available for researchers through a standard API. (Continues planning work carried out this semester.)
• There will not be an independent research project in information retrieval.
3
Course Administration
Final Examination
• The final examination is on Monday, December 13, between 12:00 and 1:30.
• There appears to be little interest in a make-up examination on another date. If, however, you have real problems with the scheduled date, send an email message to Anat Nidar-Levi ([email protected]).
4
Cluster Analysis
Cluster Analysis
Methods that divide a set of n objects into m non-overlapping subsets.
For information discovery, cluster analysis is applied to
• terms for thesaurus construction
• documents to divide into categories (sometimes called automatic classification, but classification usually requires a pre-determined set of categories).
5
Cluster Analysis Metrics
Documents clustered on the basis of a similarity measure calculated from the terms that they contain.
Documents clustered on the basis of co-occurring citations.
Terms clustered on the basis of the documents in which they co-occur.
6
Non-hierarchical and Hierarchical Methods
Non-hierarchical methods
Elements are divided into m non-overlapping sets where m is predetermined.
Hierarchical methods
m is varied progressively to create a hierarchy of solutions.
Agglomerative methods
m is initially equal to n, the total number of elements, where every element is considered to be a cluster with one element.
The hierarchy is produced by incrementally combining clusters.
7
Simple methods: Single Link
x
xx
xx
xxx
x
x
x
x
Similarity between clusters is similarity between most similar elements
Concept
8
Simple methods: Single Link
Single Link
A simple agglomerative method.
Initially, each element is its own cluster with one element.
At each step, calculate the similarity between each pair of clusters as the most similar pair of elements that are not yet in the same cluster. Merge the two clusters that are most similar.
May lead to long, straggling clusters (chaining).
Very simple computation.
9
Similarities: Incidence array
D1: alpha bravo charlie delta echo foxtrot golf
D2: golf golf golf delta alpha
D3: bravo charlie bravo echo foxtrot bravo
D4: foxtrot alpha alpha golf golf delta
alpha bravo charlie delta echo foxtrot golf
D1 1 1 1 1 1 1 1
D2 1 1 1
D3 1 1 1 1
D4 1 1 1 1
n 3 2 2 3 2 3 3
10
Term similarity matrix
alpha bravo charlie delta echo foxtrot golf
alpha 0.2 0.2 0.5 0.2 0.33 0.5
bravo 0.5 0.2 0.5 0.4 0.2
charlie 0.2 0.5 0.4 0.2
delta 0.2 0.33 0.5
echo 0.4 0.2
foxtrot 0.33
golf
Using incidence matrix and dice weighting
11
Example -- single link
alpha delta golf bravo echo charlie foxtrot
1
Agglomerative: step 1
12
Example -- single link
alpha delta golf bravo echo charlie foxtrot
1
2
Agglomerative: step 2
13
Example -- single link
alpha delta golf bravo echo charlie foxtrot
1
23
Agglomerative: step 3
14
Example -- single link
alpha delta golf bravo echo charlie foxtrot
1
23
6
4
5
This style of diagram is called a dendrogram.
15
Simple methods: Complete Linkage
x
xx
xx
xxx
x
x
x
x
Similarity between clusters is similarity between least similar elements
Concept
16
Simple methods: complete linkage
Complete linkage
A simple agglomerative method.
Initially, each element is its own cluster with one element.
At each step, calculate the similarity between each pair of clusters as the similarity between the least similar pair of elements in the two clusters. Merge the two clusters that are most similar.
Generates small, tightly bound clusters
17
Term similarity matrix
alpha bravo charlie delta echo foxtrot golf
alpha 0.2 0.2 0.5 0.2 0.33 0.5
bravo 0.5 0.2 0.5 0.4 0.2
charlie 0.2 0.5 0.4 0.2
delta 0.2 0.33 0.5
echo 0.4 0.2
foxtrot 0.33
golf
Using incidence matrix and dice weighting
18
Example – complete linkage
Cluster a b c d e f gelements
Least similar pair / distance
a - ab/.2 ac/.2 ad/.5 ae/.2 af/.33 ag/.5 b - bc/.5 bd/.2 be/.5 bf/.4 bg/.2 c - cd/.2 ce/.5 cf/.4 cg/.2 d - de/.2 df/.33 dg/.5 e - ef/.4 eg/.2 f - fg/.33 g -
Step 1. Merge clusters {a} and {d}
19
Example – complete linkage
Cluster a,d b c e f gelements
Least similar pair / distance
a,d - ab/.2 ac/.2 ae/.2 df/.33 ag/.5 b - bc/.5 be/.5 bf/.4 bg/.2 c - ce/.5 cf/.4 cg/.2 e - ef/.4 eg/.2 f - fg/.33 g -
Step 2. Merge clusters {a,d} and {g}
20
Example – complete linkage
Cluster a,d,g b c e felements
Least similar pair / distance
a,d,g - ab/.2 ac/.2 ae/.2 af/.33 b - bc/.5 be/.5 bf/.4 c - ce/.5 cf/.4 e - ef/.4 f -
Step 3. Merge clusters {b} and {c}
21
Example – complete linkage
Cluster a,d,g b,c e felements
Least similar pair / distance
a,d,g - ab/.2 ae/.2 af/.33 b,c - be/.5 bf/.4 e - ef/.4 f -
Step 4. Merge clusters {b,c} and {e}
22
Example -- complete linkage
alpha delta golf bravo charlie echo foxtrot
Step 1
Step 6Step 5
Step 2
Step 4Step 3
23
Problems with cluster analysis
Selection of attributes on which items are clustered
Choice of similarity measure and algorithm
Computational resources
Assessing validity and stability of clusters
Updating clusters as data changes
Method for searching the clusters
24
Example 1: Concept Spaces for Scientific Terms
Large-scale searches can only match terms specified by the user to terms appearing in documents. Cluster analysis can be used to provide information retrieval by concepts, rather than by terms.
Bruce Schatz, William H. Mischo, Timothy W. Cole, Joseph B. Hardin, Ann P. Bishop (University of Illinois), Hsinchun Chen (University of Arizona), Federating Diverse Collections of Scientific Literature, IEEE Computer, May 1996. Federating Diverse Collections of Scientific Literature
25
Concept Spaces: Methodology
Concept space:
A similarity matrix based on co-occurrence of terms.
Approach:
Use cluster analysis to generate "concept spaces" automatically, i.e., clusters of terms that embrace a single semantic concept.
Arrange concepts in a hierarchical classification.
26
Concept Spaces: INSPEC Data
Data set 1: All terms in 400,000 records from INSPEC, containing 270,000 terms with 4,000,000 links.
[24.5 hours of CPU on 16-node Silicon Graphics supercomputer.]
computer-aided instructionsee also educationUF teaching machinesBT educational computingTT computer applicationsRT educationRT teaching
27
Concept Space: Compendex Data
Data set 2:
(a) 4,000,000 abstracts from the Compendex database covering all of engineering as the collection, partitioned along classification code lines into some 600 community repositories.
[ Four days of CPU on 64-processor Convex Exemplar.]
(b) In the largest experiment, 10,000,000 abstracts, were divided into sets of 100,000 and the concept space for each set generated separately. The sets were selected by the existing classification scheme.
28
Objectives
• Semantic retrieval (using concept spaces for term suggestion)
• Semantic interoperability (vocabulary switching across subject domains)
• Semantic indexing (concept identification of document content)
• Information representation (information units for uniform manipulation)
29
Use of Concept Space: Term Suggestion
30
Future Use of Concept Space: Vocabulary Switching
"I'm a civil engineer who designs bridges. I'm interested in using fluid dynamics to compute the structural effects of wind currents on long structures. Ocean engineers who design undersea cables probably do similar computations for the structural effects of water currents on long structures. I want you [the system] to change my civil engineering fluid dynamics terms into the ocean engineering terms and search the undersea cable literature."
31
Example 2: Visual thesaurus for geographic images
Methodology:
• Divide images into small regions.
• Create a similarity measure based on properties of these images.
• Use cluster analysis tools to generate clusters of similar images.
• Provide alternative representations of clusters.
Marshall Ramsey, Hsinchun Chen, Bin Zhu, A Collection of Visual Thesauri for Browsing Large Collections of Geographic Images, May 1997. http://ai.bpa.arizona.edu/~mramsey/papers/visualThesaurus/visualThesaurus.html
32
33
Example 3: Cluster Analysis of Social Science Journal
In the social sciences, subject boundaries are unclear. Can citation patterns be used to develop criteria for matching information services to the interests of users?
W. Y. Arms and C. R. Arms, Cluster analysis used on social science citations, Journal of Documentation, 34 (1) pp 1-11, March 1978.
34
Methodology
Assumption: Two journals are close to each other if they are cited by the same source journals, with similar relative frequencies.
Sources of citations: Select a sample of n social science journals.
Citation matrix: Construct an m x n matrix in which the ijth element is the number of citations to journal i from journal j.
Normalization: All data was normalized so that the sum of the elements in each row is 1.
35
Data
Pilot study:
5,000 citations from the 1970 volumes of 17 major journals from across the social sciences.
Criminology citations:
Every fifth citation from a set of criminology journals (3 sets of data for 1950, 1960, 1970).
Main file (52,000 citations):
(a) Every citation from the 1970 volumes of the 48 most cited source journals in the pilot study.
(b) Every citation from the 1970 volumes of 47 randomly selected journals.
36
Sample sizes
Sample Source journals Target journals
Pilot 17 115
Criminology: 1950 10 18 1960 13 49 1970 27 108
Main file: ranked 48 495 random 47 254
Excludes journals that are cited by only one source. These were assumed to cluster with the source.
37
Algorithms
Main analysis used a non-hierarchical method of E. M. L. Beale and M. G. Kendal based on Euclidean distance.
For comparison, 36 psychology journals clustered using:
single-linkagecomplete-linkagevan Rijsbergen's algorithm
Beale/Kendal algorithm and complete-linkage produced similar results.
Single-linkage suffered from chaining.
Van Rijsbergen algorithm seeks very clear-cut clusters, which were not found in the data.
38
Non-hierarchical clusters
Economics clusters in the pilot study
39
Non-hierarchical dendrogram
Part of a dendrogram showing non-hierarchical structure
40
Conclusion
"The overall conclusion must be that cluster analysis is not a practical method of designing secondary services in the social sciences."
• Because of skewed distributions very large amounts of data are required.
• Results are complex and difficult to interpret.
• Overlap between social sciences leads to results that are sensitive to the precise data and algorithms chosen.
41
The End
Search index
Return hits
Browse content
Return objects