Concepts and Techniques: - Chapter 10

Data Mining:
Concepts and
Techniques
(3rd ed.)
Chapter 10
Jiawei Han, Micheline Kamber, and Jian Pei
University of Illinois at Urbana-Champaign &
Simon Fraser University
2013 Han, Kamber & Pei. All rights reserved.
1
Chapter 10. Cluster Analysis: Basic

Concepts and Methods
Cluster Analysis: Basic Concepts
Partitioning Methods
Hierarchical Methods
Density-Based Methods
Grid-Based Methods
Evaluation of Clustering
Summary
3
What is Cluster Analysis?
Cluster: A collection of data objects

similar (or related) to one another within the same group
dissimilar (or unrelated) to the objects in other groups
Cluster analysis (or clustering, data segmentation, )
Finding similarities between data according to the
characteristics found in the data and grouping similar
data objects into clusters
Unsupervised learning: no predefined classes (i.e., learning
by observations vs. learning by examples: supervised)
Typical applications
As a stand-alone tool to get insight into data distribution
As a preprocessing step for other algorithms
Applications of Cluster Analysis
Data reduction
Summarization: Preprocessing for regression, PCA,

classification, and association analysis
Compression: Image processing: vector quantization

Hypothesis generation and testing
Prediction based on groups
Cluster & find characteristics/patterns for each group
Finding K-nearest Neighbors
Localizing search to one or a small number of clusters
Outlier detection: Outliers are often viewed as those far

away from any cluster
Clustering: Application Examples
Biology: taxonomy of living things: kingdom, phylum, class,

order, family, genus and species
Information retrieval: document clustering
Land use: Identification of areas of similar land use in an
earth observation database
Marketing: Help marketers discover distinct groups in their
customer bases, and then use this knowledge to develop
targeted marketing programs
City-planning: Identifying groups of houses according to their
house type, value, and geographical location
Earth-quake studies: Observed earth quake epicenters
should be clustered along continent faults
Climate: understanding earth climate, find patterns of
atmospheric and ocean
Economic Science: market resarch
6
Basic Steps to Develop a Clustering

Task
Feature selection
Select info concerning the task of interest
Minimal information redundancy
Proximity measure
Clustering criterion
Choice of algorithms
Validation of the results
Expressed via a cost function or some rules
Clustering algorithms
Similarity of two feature vectors
Validation test (also, clustering tendency test)
Interpretation of the results
Integration with applications

7
Quality: What Is Good Clustering?
A good clustering method will produce high quality

clusters
high intra-class similarity: cohesive within clusters
low inter-class similarity: distinctive between

clusters
The quality of a clustering method depends on
the similarity measure used by the method
its implementation, and
Its ability to discover some or all of the hidden

patterns
8
Measure the Quality of Clustering
Dissimilarity/Similarity metric
Similarity is expressed in terms of a distance

function, typically metric: d(i, j)
The definitions of distance functions are usually

rather different for interval-scaled, boolean,
categorical, ordinal ratio, and vector variables
Weights should be associated with different

variables based on applications and data semantics
Quality of clustering:
There is usually a separate quality function that

measures the goodness of a cluster.
It is hard to define similar enough or good

enough
The answer is typically highly subjective

9
Considerations for Cluster

Analysis
Partitioning criteria
Separation of clusters
Exclusive (e.g., one customer belongs to only one region)

vs. non-exclusive (e.g., one document may belong to more
than one class)
Similarity measure
Single level vs. hierarchical partitioning (often, multi-level

hierarchical partitioning is desirable)
Distance-based (e.g., Euclidian, road network, vector) vs.

connectivity-based (e.g., density or contiguity)
Clustering space
Full space (often when low dimensional) vs. subspaces

(often in high-dimensional clustering)
10
Requirements and Challenges
Scalability
Clustering all the data instead of only on samples
Ability to deal with different types of attributes
Numerical, binary, categorical, ordinal, linked, and mixture
of these
Constraint-based clustering
User may give inputs on constraints
Use domain knowledge to determine input parameters

Interpretability and usability
Others
Discovery of clusters with arbitrary shape
Ability to deal with noisy data
Incremental clustering and insensitivity to input order
High dimensionality
11
Major Clustering Approaches (I)
Partitioning approach:
Construct various partitions and then evaluate them by
some criterion, e.g., minimizing the sum of square errors
Typical methods: k-means, k-medoids, CLARANS
Hierarchical approach:
Create a hierarchical decomposition of the set of data (or
objects) using some criterion
Typical methods: Diana, Agnes, BIRCH, CAMELEON
Density-based approach:
Based on connectivity and density functions
Typical methods: DBSACN, OPTICS, DenClue
Grid-based approach:
based on a multiple-level granularity structure
Typical methods: STING, WaveCluster, CLIQUE
12
Major Clustering Approaches (II)
Model-based:
A model is hypothesized for each of the clusters and tries to
find the best fit of that model to each other
Typical methods: EM, SOM, COBWEB
Frequent pattern-based:
Based on the analysis of frequent patterns
Typical methods: p-Cluster
User-guided or constraint-based:
Clustering by considering user-specified or applicationspecific constraints
Typical methods: COD (obstacles), constrained clustering
Link-based clustering:
Objects are often linked together in various ways
Massive links can be used to cluster objects: SimRank,
LinkClus
13

Grid-Based Methods
Summary
14
Partitioning Algorithms: Basic

Concept
Partitioning method: Partitioning a database D of n objects into

a set of k clusters, such that the sum of squared distances is
minimized (where ci is the centroid or medoid of cluster C i)
E ik1 pCi (d ( p, ci )) 2
Given k, find a partition of k clusters that optimizes the chosen

partitioning criterion
Global optimal: exhaustively enumerate all partitions
Heuristic methods: k-means and k-medoids algorithms
k-means (MacQueen67, Lloyd57/82): Each cluster is

represented by the center of the cluster
k-medoids or PAM (Partition around medoids) (Kaufman &

Rousseeuw87): Each cluster is represented by one of the
objects in the cluster
15
The K-Means Clustering

Method
Given k, the k-means algorithm is implemented in

four steps:
Partition objects into k nonempty subsets
Compute seed points as the centroids of the

clusters of the current partitioning (the centroid
is the center, i.e., mean point, of the cluster)
Assign each object to the cluster with the

nearest seed point
Go back to Step 2, stop when the assignment

does not change
16
An Example of K-Means Clustering

K=2
The initial data

set
Arbitrarily
partition
objects
into k
groups
Partition objects into k nonempty

subsets
Repeat
Compute centroid (i.e., mean

point) for each partition
Assign each object to the

cluster of its nearest centroid
Update
the
cluster
centroids
Loop if
needed
Reassign objects
Update
the
cluster
centroids
Until no change
17
Comments on the K-Means

Method
Strength: Efficient: O(tkn), where n is # objects, k is # clusters, and

t is # iterations. Normally, k, t << n.
Comparing: PAM: O(k(n-k)2 ), CLARA: O(ks2 + k(n-k))
Comment: Often terminates at a local optimal
Weakness
Applicable only to objects in a continuous n-dimensional space
Using the k-modes method for categorical data

In comparison, k-medoids can be applied to a wide range of
data
Need to specify k, the number of clusters, in advance (there are

ways to automatically determine the best k (see Hastie et al.,
2009)
Sensitive to noisy data and outliers
Not suitable to discover clusters with non-convex shapes

18
Variations of the K-Means Method
Most of the variants of the k-means which differ in
Selection of the initial k means
Dissimilarity calculations
Strategies to calculate cluster means
Handling categorical data: k-modes
Replacing means of clusters with modes
Using new dissimilarity measures to deal with categorical

objects
Using a frequency-based method to update modes of clusters
A mixture of categorical and numerical data: k-prototype

method
19
What Is the Problem of the K-Means Method?
The k-means algorithm is sensitive to outliers !
Since an object with an extremely large value may

substantially distort the distribution of the data
K-Medoids: Instead of taking the mean value of the object in

a cluster as a reference point, medoids can be used, which is
the most centrally located object in a cluster
10
10
0
0
10
10
20
PAM: A Typical K-Medoids Algorithm
Total Cost = 20
10
10
10
6
5
4
3
2
1
0
0
K=2
10
Arbitrar
y
choose
k object
as
initial
medoid
s
6
5
4
3
2
1
0
0
10
Total Cost = 26
10
Do loop
Until no
change
Assign
each
remaini
ng
object
to
nearest
medoid
s
7
6
5
4
3
2
1
0
0
10
Compute
total cost
of
swapping
Swapping
O and
Oramdom
If quality is
improved.
6
5
4
10
Randomly select a
nonmedoid
object,Oramdom
8
7
8
7
6
5
4
0
0
10
10
21
The K-Medoid Clustering Method
K-Medoids Clustering: Find representative objects (medoids) in clusters
PAM (Partitioning Around Medoids, Kaufmann & Rousseeuw 1987)
Starts from an initial set of medoids and iteratively replaces one of

the medoids by one of the non-medoids if it improves the total
distance of the resulting clustering
PAM works effectively for small data sets, but does not scale well
for large data sets (due to the computational complexity)
Efficiency improvement on PAM
CLARA (Kaufmann & Rousseeuw, 1990): PAM on samples
CLARANS (Ng & Han, 1994): Randomized re-sampling
22

Grid-Based Methods
Summary
23
Hierarchical Clustering
Use distance matrix as clustering criteria. This

method does not require the number of clusters k
as an input, but needs a termination condition
Step 0
a
Step 1
Step 2 Step 3 Step 4
ab
abcde
cde
de
e
Step 4
agglomerative
(AGNES)
Step 3
Step 2 Step 1 Step 0
divisive
(DIANA)
24
AGNES (Agglomerative
Nesting)
Introduced in Kaufmann and Rousseeuw (1990)
Implemented in statistical packages, e.g., Splus
Use the single-link method and the dissimilarity matrix
Merge nodes that have the least dissimilarity
Go on in a non-descending fashion
Eventually all nodes belong to the same cluster

10
10
10
0
0
10
0
0
10
10
25
Dendrogram: Shows How Clusters are

Merged
Decompose data objects into a several levels of nested partitioning (tree of
clusters), called a dendrogram
A clustering of the data objects is obtained by cutting the dendrogram at
the desired level, then each connected component forms a cluster
26
DIANA (Divisive Analysis)
Introduced in Kaufmann and Rousseeuw (1990)
Implemented in statistical analysis packages, e.g.,

Splus
Inverse order of AGNES
Eventually each node forms a cluster on its own

10
10
10
0
0
10
0
0
10
10
27
Distance between
Clusters
Single link: smallest distance between an element in one

cluster and an element in the other, i.e., dist(K i, Kj) = min(tip, tjq)
Complete link: largest distance between an element in one

cluster and an element in the other, i.e., dist(K i, Kj) = max(tip, tjq)
Average: avg distance between an element in one cluster and

an element in the other, i.e., dist(K i, Kj) = avg(tip, tjq)
Centroid: distance between the centroids of two clusters, i.e.,

dist(Ki, Kj) = dist(Ci, Cj)
Medoid: distance between the medoids of two clusters, i.e.,

dist(Ki, Kj) = dist(Mi, Mj)
Medoid: a chosen, centrally located object in the cluster

28
Centroid, Radius and Diameter

of a Cluster (for numerical
data sets)
Centroid: the middle of a cluster
Cm
iN 1(t
ip
Radius: square root of average distance from any

point of the cluster to its centroid
N (t cm ) 2
Rm i 1 ip
N
Diameter: square root of average mean squared
distance between all pairs of points in the cluster

N N (t t ) 2
Dm i 1 i 1 ip iq
N ( N 1)
29
Extensions to Hierarchical
Clustering
Major weakness of agglomerative clustering methods
Can never undo what was done previously
Do not scale well: time complexity of at least O(n2),

where n is the number of total objects
Integration of hierarchical & distance-based clustering
BIRCH (1996): uses CF-tree and incrementally

adjusts the quality of sub-clusters
CHAMELEON (1999): hierarchical clustering using

dynamic modeling
30
Reducing and Clustering Using

Hierarchies)
Zhang, Ramakrishnan & Livny, SIGMOD96
Incrementally construct a CF (Clustering Feature) tree, a

hierarchical data structure for multiphase clustering
Phase 1: scan DB to build an initial in-memory CF tree (a

multi-level compression of the data that tries to preserve
the inherent clustering structure of the data)
Phase 2: use an arbitrary clustering algorithm to cluster the

leaf nodes of the CF-tree
Scales linearly: finds a good clustering with a single scan and

improves the quality with a few additional scans
Weakness: handles only numeric data, and sensitive to the

order of the data record
31
Clustering Feature Vector in BIRCH

Clustering Feature (CF): CF = (N, LS, SS)
N: Number of data points
LS: linear sum of N points:
Xi
i 1
SS: square sum of N points
CF = (5, (16,30),(54,190))
N
Xi
i 1
10
(3,4)
(2,6)
(4,5)
(4,7)
(3,8)
9
8
7
6
5
4
3
2
1
0
0
32
10
CF-Tree in BIRCH
Clustering feature:
Summary of the statistics for a given subcluster: the 0th, 1st, and 2nd moments of the subcluster from the
statistical point of view
Registers crucial measurements for computing cluster

and utilizes storage efficiently
A CF tree is a height-balanced tree that stores the

clustering features for a hierarchical clustering
A nonleaf node in a tree has descendants or children
The nonleaf nodes store sums of the CFs of their children
A CF tree has two parameters
Branching factor: max # of children
Threshold: max diameter of sub-clusters stored at the

leaf nodes
33
The CF Tree Structure

Root
B=7
CF1
CF2
CF3
CF6
L=6
child1
child2
child3
child6
Non-leaf node
CF1
CF2
CF3
CF5
child1
child2
child3
child5
Leaf node
prev CF1
CF2
CF6
Leaf node
next
prev CF1
CF2
CF4
next
34
The Birch Algorithm
Cluster Diameter
For each point in the input

Find closest leaf entry
Add point to leaf entry and update CF
If entry diameter > max_diameter, then split leaf, and
possibly parents
Algorithm is O(n)
Concerns
Sensitive to insertion order of data points
Since we fix the size of leaf nodes, so clusters may not be so
natural
Clusters tend to be spherical given the radius and diameter
measures
1
2
( xi x j )
n( n 1)
35
CHAMELEON: Hierarchical Clustering

Using Dynamic Modeling (1999)
CHAMELEON: G. Karypis, E. H. Han, and V. Kumar, 1999
Measures the similarity based on a dynamic model
Two clusters are merged only if the interconnectivity

and closeness (proximity) between two clusters are
high relative to the internal interconnectivity of the
clusters and closeness of items within the clusters
Graph-based, and a two-phase algorithm

1.
2.
Use a graph-partitioning algorithm: cluster objects into

a large number of relatively small sub-clusters
Use an agglomerative hierarchical clustering
algorithm: find the genuine clusters by repeatedly
combining these sub-clusters
36
KNN Graphs & Interconnectivity
k-nearest graphs from an original data in 2D:
EC{Ci ,Cj } :The absolute inter-connectivity between Ci and

Cj: the sum of the weight of the edges that connect
vertices in Ci to vertices in Cj
Internal inter-connectivity of a cluster Ci : the size of its

min-cut bisector ECCi (i.e., the weighted sum of edges
that partition the graph into two roughly equal parts)
Relative Inter-connectivity (RI):
37
Relative Closeness & Merge of SubClusters
Relative closeness between a pair of clusters Ci and

Cj : the absolute closeness between Ci and Cj
normalized w.r.t. the internal closeness of the two
clusters Ci and Cj
and
are the average weights of the edges that
belong in the min-cut bisector of clusters Ci and Cj ,
respectively, and
is the average weight of the edges
that connect vertices in Ci to vertices in Cj
Merge Sub-Clusters:
Merges only those pairs of clusters whose RI and RC are both

above some user-specified thresholds
Merge those maximizing the function that combines RI and RC
38
Overall Framework of
CHAMELEON
Construct (K-NN)
Partition the Graph
Sparse Graph
Data Set
K-NN Graph
P and q are
connected if q is
among the top k
closest neighbors of
p
Merge Partition
Final Clusters
Relative
interconnectivity:
connectivity of c1 and
c2 over internal
connectivity
Relative closeness:
39
CHAMELEON (Clustering Complex

Objects)
40
Probabilistic Hierarchical
Clustering
Algorithmic hierarchical clustering
Nontrivial to choose a good distance measure
Hard to handle missing attribute values
Optimization goal not clear: heuristic, local search
Probabilistic hierarchical clustering
Use probabilistic models to measure distances between clusters
Generative model: Regard the set of data objects to be

clustered as a sample of the underlying data generation
mechanism to be analyzed
Easy to understand, same efficiency as algorithmic

agglomerative clustering method, can handle partially observed
data
In practice, assume the generative models adopt common

distribution functions, e.g., Gaussian distribution or Bernoulli
distribution, governed by parameters
41
Generative Model
Given a set of 1-D points X = {x1, , xn} for

clustering analysis & assuming they are
generated by a Gaussian distribution:
The probability that a point xi X is generated

by the model
The likelihood that X is generated by the model:
The task of learning the generative model: find

the maximum
the parameters and 2 such that
likelihood
42
Gaussian Distribution
Bean
machine:
drop ball
with pins
1-d
Gaussia
n
From wikipedia and

http://home.dei.polimi.it
2-d
Gaussia
n
43
A Probabilistic Hierarchical Clustering

Algorithm
For a set of objects partitioned into m clusters C1, . . . ,Cm, the

quality can be measured by,
where P() is the maximum likelihood

If we merge two clusters C j1 and Cj2 into a cluster Cj1Cj2, then,
the change in quality of the overall clustering is
Distance between clusters C1 and C2:
If dist(Ci, Cj) < 0, merge Ci and Cj

44

Grid-Based Methods
Summary
45
Density-Based Clustering
Methods
Clustering based on density (local cluster criterion),

such as density-connected points
Major features:
Discover clusters of arbitrary shape
Handle noise
One scan
Need density parameters as termination condition
Several interesting studies:
DBSCAN: Ester, et al. (KDD96)
OPTICS: Ankerst, et al (SIGMOD99).
DENCLUE: Hinneburg & D. Keim (KDD98)
CLIQUE: Agrawal, et al. (SIGMOD98) (more gridbased)
46
Density-Based Clustering: Basic

Concepts
Two parameters:
Eps: Maximum radius of the neighbourhood
MinPts: Minimum number of points in an Epsneighbourhood of that point
NEps(q): {p belongs to D | dist(p,q) Eps}
Directly density-reachable: A point p is directly

density-reachable from a point q w.r.t. Eps,
MinPts if
p belongs to NEps(q)
core point condition:
MinPts = 5
Eps = 1 cm
|NEps (q)| MinPts

47
Density-Reachable and DensityConnected
Density-reachable:
A point p is density-reachable
from a point q w.r.t. Eps, MinPts if
there is a chain of points p1, ,
pn, p1 = q, pn = p such that pi+1 is
directly density-reachable from pi
Density-connected
A point p is density-connected to
a point q w.r.t. Eps, MinPts if there
is a point o such that both, p and
q are density-reachable from o
w.r.t. Eps and MinPts
p
p1
q
o
48
DBSCAN: Density-Based Spatial

Clustering of Applications with
Noise
Relies on a density-based notion of cluster: A

cluster is defined as a maximal set of densityconnected points
Discovers clusters of arbitrary shape in spatial
databases with noise
Outlier
Border
Eps = 1cm
Core
MinPts = 5
49
DBSCAN: The Algorithm
Arbitrary select a point p
Retrieve all points density-reachable from p w.r.t. Eps

and MinPts
If p is a core point, a cluster is formed
If p is a border point, no points are density-reachable

from p and DBSCAN visits the next point of the database
Continue the process until all of the points have been

processed
If a spatial index is used, the computational complexity of DBSCAN

is O(nlogn), where n is the number of database objects. Otherwise,
the complexity is O(n2)
50
DBSCAN: Sensitive to Parameters
DBSCAN online Demo:

http://webdocs.cs.ualberta.ca/~yaling/Cluster/Applet/Code/Clus
51
OPTICS: A Cluster-Ordering Method

(1999)
OPTICS: Ordering Points To Identify the Clustering

Structure
Ankerst, Breunig, Kriegel, and Sander (SIGMOD99)
Produces a special order of the database wrt its
density-based clustering structure
This cluster-ordering contains info equiv to the
density-based clusterings corresponding to a
broad range of parameter settings
Good for both automatic and interactive cluster
analysis, including finding intrinsic clustering
structure
Can be represented graphically or using
visualization techniques
52
OPTICS: Some Extension from DBSCAN
Index-based: k = # of dimensions, N: # of points

Complexity: O(N*logN)
Core Distance of an object p: the smallest value such
that the -neighborhood of p has at least MinPts objects
Let N(p): -neighborhood of p, is a distance value
Core-distance, MinPts(p) = Undefined if card(N(p)) < MinPts
MinPts-distance(p), otherwise
Reachability Distance of object p from core object q is the
min radius value that makes p density-reachable from q
Reachability-distance, MinPts(p, q) =
Undefined if q is not a core object
max(core-distance(q), distance (q, p)), otherwise
53
Core Distance & Reachability

Distance
54
Reachabilitydistance
undefined
Cluster-order of the objects

55
Density-Based Clustering: OPTICS &

Applications
demo:
http://www.dbs.informatik.uni-muenchen.de/Forschung/KDD/Clustering/OPTICS/Demo
56
DENCLUE: Using Statistical Density

Functions
DENsity-based CLUstEring by Hinneburg & Keim (KDD98)
Using statistical density functions:
f Gaussian ( x , y ) e
d ( x,y)
2 2
influence of
y on x
Major features
f
f
total influence
on x
D
Gaussian
D
Gaussian
( x ) i 1 e
N
d ( x , xi ) 2
2
( x, xi ) i 1 ( xi x) e
N
d ( x , xi ) 2
2 2
gradient of x
in the
direction of xi
Solid mathematical foundation
Good for data sets with large amounts of noise
Allows a compact mathematical description of arbitrarily

shaped clusters in high-dimensional data sets
Significant faster than existing algorithm (e.g., DBSCAN)
But needs a large number of parameters

57
Denclue: Technical Essence
Uses grid cells but only keeps information about grid cells that
do actually contain data points and manages these cells in a
tree-based access structure
Influence function: describes the impact of a data point within
its neighborhood
Overall density of the data space can be calculated as the
sum of the influence function of all data points
Clusters can be determined mathematically by identifying
density attractors
Density attractors are local maximal of the overall density
function
Center defined clusters: assign to each density attractor the
points density attracted to it
Arbitrary shaped cluster: merge density attractors that are
connected through paths of high density (> threshold)
58
Density Attractor
59
Center-Defined and Arbitrary
60

Grid-Based Methods
Summary
61
Grid-Based Clustering Method
Using multi-resolution grid data structure

Several interesting methods
STING (a STatistical INformation Grid
approach) by Wang, Yang and Muntz (1997)
CLIQUE: Agrawal, et al. (SIGMOD98)
Both grid-based and subspace clustering
WaveCluster by Sheikholeslami, Chatterjee,

and Zhang (VLDB98)
A multi-resolution clustering approach

using wavelet method
62
STING: A Statistical Information Grid

Approach
Wang, Yang and Muntz (VLDB97)

The spatial area is divided into rectangular cells
There are several levels of cells corresponding to different
levels of resolution
63
The STING Clustering Method
Each cell at a high level is partitioned into a number of

smaller cells in the next lower level
Statistical info of each cell is calculated and stored
beforehand and is used to answer queries
Parameters of higher level cells can be easily calculated
from parameters of lower level cell
count, mean, s, min, max
type of distributionnormal, uniform, etc.
Use a top-down approach to answer spatial data queries
Start from a pre-selected layertypically with a small
number of cells
For each cell in the current level compute the
confidence interval
64
STING Algorithm and Its

Analysis
Remove the irrelevant cells from further

consideration
When finish examining the current layer, proceed to
the next lower level
Repeat this process until the bottom layer is reached
Advantages:
Query-independent, easy to parallelize,
incremental update
O(K), where K is the number of grid cells at the
lowest level
Disadvantages:
All the cluster boundaries are either horizontal or
vertical, and no diagonal boundary is detected
65
CLIQUE (Clustering In QUEst)
Agrawal, Gehrke, Gunopulos, Raghavan (SIGMOD98)
Automatically identifying subspaces of a high dimensional data

space that allow better clustering than original space
CLIQUE can be considered as both density-based and grid-based
It partitions each dimension into the same number of equal

length interval
It partitions an m-dimensional data space into nonoverlapping rectangular units
A unit is dense if the fraction of total data points contained in

the unit exceeds the input model parameter
A cluster is a maximal set of connected dense units within a

subspace
66
CLIQUE: The Major Steps
Partition the data space and find the number of points

that lie inside each cell of the partition.
Identify the subspaces that contain clusters using the

Apriori principle
Identify clusters
Determine dense units in all subspaces of interests

Determine connected dense units in all subspaces
of interests.
Generate minimal description for the clusters

Determine maximal regions that cover a cluster of
connected dense units for each cluster
Determination of minimal cover for each cluster
67
=3
30
40
Vacation
20
50
Salary
(10,000)
0 1 2 3 4 5 6 7
la
a
S
ry
30
Vacation(
week)
0 1 2 3 4 5 6 7
age
60
20
50
30
40
50
age
60
age
68
Strength and Weakness of CLIQUE
Strength
automatically finds subspaces of the highest
dimensionality such that high density clusters exist
in those subspaces
insensitive to the order of records in input and does
not presume some canonical data distribution
scales linearly with the size of input and has good
scalability as the number of dimensions in the data
increases
Weakness
The accuracy of the clustering result may be
degraded at the expense of simplicity of the method
69

Grid-Based Methods
Summary
70
Determine the Number of Clusters
Empirical method
# of clusters: k n/2 for a dataset of n points, e.g., n = 200, k = 10
Elbow method
Use the turning point in the curve of sum of within cluster variance
w.r.t the # of clusters
Cross validation method
Divide a given data set into m parts
Use m 1 parts to obtain a clustering model
Use the remaining part to test the quality of the clustering
E.g., For each point in the test set, find the closest centroid, and
use the sum of squared distance between all points in the test
set and the closest centroids to measure how well the model fits
the test set
For any k > 0, repeat it m times, compare the overall quality
measure w.r.t. different ks, and find # of clusters that fits the data
the best
71
Measuring Clustering Quality
3 kinds of measures: External, internal and relative
External: supervised, employ criteria not inherent to the

dataset
Internal: unsupervised, criteria derived from data itself
Compare a clustering against prior or expert-specified

knowledge (i.e., the ground truth) using certain clustering
quality measure
Evaluate the goodness of a clustering by considering how
well the clusters are separated, and how compact the
clusters are, e.g., Silhouette coefficient
Relative: directly compare different clusterings, usually those

obtained via different parameter settings for the same
algorithm
72
Measuring Clustering Quality: External

Methods
Clustering quality measure: Q(C, T), for a clustering C

given the ground truth T
Q is good if it satisfies the following 4 essential criteria
Cluster homogeneity: the purer, the better
Cluster completeness: should assign objects belong
to the same category in the ground truth to the same
cluster
Rag bag: putting a heterogeneous object into a pure
cluster should be penalized more than putting it into
a rag bag (i.e., miscellaneous or other category)
Small cluster preservation: splitting a small category
into pieces is more harmful than splitting a large
category into pieces
73
Some Commonly Used External Measures
Matching-based measures
Purity, maximum matching, F-measure
Entropy-Based Measures
Ground truth partitioning T T
Conditional entropy, normalized mutual
Cluster
Cluster
C
information (NMI), variation of information C
Pair-wise measures
Four possibilities: True positive (TP), FN, FP,
TN
Jaccard coefficient, Rand statistic, FowlkesMallow measure
Correlation measures
Discretized Huber static, normalized
discretized Huber static
1
74
Entropy-Based Measure (I):

Conditional Entropy
Entropy of clustering C:
Entropy of partitioning T:
Entropy of T w.r.t. cluster Ci:
Conditional entropy of T
w.r.t. clustering C:
The more a clusters members are split into different
partitions, the higher the conditional entropy
For a perfect clustering, the conditional entropy value
is 0, where the worst possible conditional entropy value
is log k
75
Entropy-Based Measure (II):
Normalized mutual information

(NMI)
Mutual information: quantify the amount of shared info

between the clustering C and partitioning T:
It measures the dependency between the observed joint

probability pij of C and T, and the expected joint probability p Ci * pTj
under the independence assumption
When C and T are independent, p ij = pCi * pTj, I(C, T) = 0. However,
there is no upper bound on the mutual information
Normalized mutual information (NMI)
Value range of NMI: [0,1]. Value close to 1 indicates a good

clustering
76

Grid-Based Methods
Summary
77
Summary
Cluster analysis groups objects based on their similarity and

has wide applications
Measure of similarity can be computed for various types of data
Clustering algorithms can be categorized into partitioning
methods, hierarchical methods, density-based methods, gridbased methods, and model-based methods
K-means and K-medoids algorithms are popular partitioningbased clustering algorithms
Birch and Chameleon are interesting hierarchical clustering
algorithms, and there are also probabilistic hierarchical
clustering algorithms
DBSCAN, OPTICS, and DENCLU are interesting density-based
algorithms
STING and CLIQUE are grid-based methods, where CLIQUE is
also a subspace clustering algorithm
Quality of clustering results can be evaluated in various ways
78
79
80
CS512-Spring 2011: An
Introduction
Coverage
Cluster Analysis: Chapter 11
Outlier Detection: Chapter 12
Mining Sequence Data: BK2: Chapter 8
Mining Graphs Data: BK2: Chapter 9
Social and Information Network Analysis
BK2: Chapter 9
Partial coverage: Mark Newman: Networks: An Introduction, Oxford U.,
2010
Scattered coverage: Easley and Kleinberg, Networks, Crowds, and
Markets: Reasoning About a Highly Connected World, Cambridge U., 2010
Recent research papers
Mining Data Streams: BK2: Chapter 8
Requirements
One research project
One class presentation (15 minutes)
Two homeworks (no programming assignment)
Two midterm exams (no final exam)

81
References (1)
R. Agrawal, J. Gehrke, D. Gunopulos, and P. Raghavan. Automatic

subspace clustering of high dimensional data for data mining
applications. SIGMOD'98
M. R. Anderberg. Cluster Analysis for Applications. Academic Press, 1973.
M. Ankerst, M. Breunig, H.-P. Kriegel, and J. Sander. Optics: Ordering
points to identify the clustering structure, SIGMOD99.
Beil F., Ester M., Xu X.: "Frequent Term-Based Text Clustering", KDD'02
M. M. Breunig, H.-P. Kriegel, R. Ng, J. Sander. LOF: Identifying DensityBased Local Outliers. SIGMOD 2000.
M. Ester, H.-P. Kriegel, J. Sander, and X. Xu. A density-based algorithm for
discovering clusters in large spatial databases. KDD'96.
M. Ester, H.-P. Kriegel, and X. Xu. Knowledge discovery in large spatial
databases: Focusing techniques for efficient class identification. SSD'95.
D. Fisher. Knowledge acquisition via incremental conceptual clustering.
Machine Learning, 2:139-172, 1987.
D. Gibson, J. Kleinberg, and P. Raghavan. Clustering categorical data: An
approach based on dynamic systems. VLDB98.
V. Ganti, J. Gehrke, R. Ramakrishan. CACTUS Clustering Categorical Data
Using Summaries. KDD'99.
82
References (2)
D. Gibson, J. Kleinberg, and P. Raghavan. Clustering categorical data:

An approach based on dynamic systems. In Proc. VLDB98.
S. Guha, R. Rastogi, and K. Shim. Cure: An efficient clustering
algorithm for large databases. SIGMOD'98.
S. Guha, R. Rastogi, and K. Shim. ROCK: A robust clustering
algorithm for categorical attributes. In ICDE'99, pp. 512-521,
Sydney, Australia, March 1999.
A. Hinneburg, D.l A. Keim: An Efficient Approach to Clustering in
Large Multimedia Databases with Noise. KDD98.
A. K. Jain and R. C. Dubes. Algorithms for Clustering Data. Printice
Hall, 1988.
G. Karypis, E.-H. Han, and V. Kumar. CHAMELEON: A Hierarchical
Clustering Algorithm Using Dynamic Modeling. COMPUTER, 32(8):
68-75, 1999.
L. Kaufman and P. J. Rousseeuw. Finding Groups in Data: an
Introduction to Cluster Analysis. John Wiley & Sons, 1990.
E. Knorr and R. Ng. Algorithms for mining distance-based outliers in
large datasets. VLDB98.
83
References (3)
G. J. McLachlan and K.E. Bkasford. Mixture Models: Inference and Applications

to Clustering. John Wiley and Sons, 1988.
R. Ng and J. Han. Efficient and effective clustering method for spatial data
mining. VLDB'94.
L. Parsons, E. Haque and H. Liu, Subspace Clustering for High Dimensional
Data: A Review, SIGKDD Explorations, 6(1), June 2004
E. Schikuta. Grid clustering: An efficient hierarchical clustering method for
very large data sets. Proc. 1996 Int. Conf. on Pattern Recognition,.
G. Sheikholeslami, S. Chatterjee, and A. Zhang. WaveCluster: A multiresolution clustering approach for very large spatial databases. VLDB98.
A. K. H. Tung, J. Han, L. V. S. Lakshmanan, and R. T. Ng. Constraint-Based
Clustering in Large Databases, ICDT'01.
A. K. H. Tung, J. Hou, and J. Han. Spatial Clustering in the Presence of
Obstacles, ICDE'01
H. Wang, W. Wang, J. Yang, and P.S. Yu.Clustering by pattern similarity in large
data sets, SIGMOD 02.
W. Wang, Yang, R. Muntz, STING: A Statistical Information grid Approach to
Spatial Data Mining, VLDB97.
T. Zhang, R. Ramakrishnan, and M. Livny. BIRCH : An efficient data clustering
method for very large databases. SIGMOD'96.
Xiaoxin Yin, Jiawei Han, and Philip Yu,
LinkClus: Efficient Clustering via Heterogeneous Semantic Links , in Proc.
2006 Int. Conf. on Very Large Data Bases (VLDB'06), Seoul, Korea, Sept. 2006.
84


What Is Cluster Analysis?
What is Good Clustering? Measuring the Quality of Clustering
Major categories of clustering methods

Clustering structures
Calculating Distance between Clusters

k-Means: A Classical Partitioning Method
Alternative Methods: k-Medoids, k-Median, and its Variations

Agglomerative and Divisive Hierarchical Clustering
BIRCH: A Hierarchical, Micro-Clustering Approach
Chameleon: A Hierarchical Clustering Algorithm Using Dynamic Modeling

DBSCAN and OPTICS: Density-Based Clustering Based on Connected Regions
DENCLUE: Clustering Based on Density Distribution Functions

Link-Based Cluster Analysis
SimRank: Exploring Links in Cluster Analysis
LinkClus: Scalability in Link-Based Cluster Analysis

Grid-Based Methods
STING: STatistical INformation Grid
WaveCluster: Clustering Using Wavelet Transformation
CLIQUE: A Dimension-Growth Subspace Clustering Method

Summary
85
85
Slides unused in class
86
A Typical K-Medoids Algorithm (PAM)
Total Cost = 20
10
10
10
6
5
4
3
2
1
0
0
K=2
10
Arbitrar
y
choose
k object
as
initial
medoid
s
6
5
4
3
2
1
0
0
10
Total Cost = 26
10
Do loop
Until no
change
Assign
each
remaini
ng
object
to
nearest
medoid
s
7
6
5
4
3
2
1
0
0
10
Compute
total cost
of
swapping
Swapping
O and
Oramdom
If quality is
improved.
6
5
4
10
Randomly select a
nonmedoid
object,Oramdom
8
7
8
7
6
5
4
0
0
10
10
87
PAM (Partitioning Around Medoids)

(1987)
PAM (Kaufman and Rousseeuw, 1987), built in Splus
Use real object to represent the cluster
Select k representative objects arbitrarily
For each pair of non-selected object h and selected

object i, calculate the total swapping cost TCih
For each pair of i and h,
If TCih < 0, i is replaced by h

Then assign each non-selected object to the
most similar representative object
repeat steps 2-3 until there is no change

88
PAM Clustering: Finding the Best

Cluster Center
Case 1: p currently belongs to oj. If oj is replaced by orandom as

a representative object and p is the closest to one of the
other representative object oi, then p is reassigned to oi
89
What Is the Problem with

PAM?
Pam is more robust than k-means in the presence of

noise and outliers because a medoid is less
influenced by outliers or other extreme values than a
mean
Pam works efficiently for small data sets but does

not scale well for large data sets.
O(k(n-k)2 ) for each iteration
where n is # of data,k is # of clusters
Sampling-based method
CLARA(Clustering LARge Applications)
90
CLARA (Clustering Large

Applications) (1990)
CLARA (Kaufmann and Rousseeuw in 1990)
Built in statistical analysis packages, such as SPlus
It draws multiple samples of the data set, applies

PAM on each sample, and gives the best
clustering as the output
Strength: deals with larger data sets than PAM
Weakness:
Efficiency depends on the sample size
A good clustering based on samples will not

necessarily represent a good clustering of the
whole data set if the sample is biased
91
CLARANS (Randomized CLARA)

(1994)
CLARANS (A Clustering Algorithm based on

Randomized Search) (Ng and Han94)
Draws sample of neighbors dynamically
The clustering process can be presented as
searching a graph where every node is a potential
solution, that is, a set of k medoids
If the local optimum is found, it starts with new
randomly selected node in search for a new local
optimum
Advantages: More efficient and scalable than both
PAM and CLARA
Further improvement: Focusing techniques and spatial
access structures (Ester et al.95)
92
ROCK: Clustering Categorical Data
ROCK: RObust Clustering using linKs

S. Guha, R. Rastogi & K. Shim, ICDE99
Major ideas
Use links to measure similarity/proximity
Not distance-based
Algorithm: sampling-based clustering
Draw random sample
Cluster with links
Label data in disk
Experiments
Congressional voting, mushroom data
93
Similarity Measure in ROCK
Traditional measures for categorical data may not work well,

e.g., Jaccard coefficient
Example: Two groups (clusters) of transactions
C1. <a, b, c, d, e>: {a, b, c}, {a, b, d}, {a, b, e}, {a, c,
d}, {a, c, e}, {a, d, e}, {b, c, d}, {b, c, e}, {b, d, e}, {c,
d, e}
C2. <a, b, f, g>: {a, b, f}, {a, b, g}, {a, f, g}, {b, f, g}
Jaccard co-efficient may lead to wrong clustering result
C1: 0.2 ({a, b, c}, {b, d, e}} to 0.5 ({a, b, c}, {a, b, d})
C1 & C2: could be as high as 0.5 ({a, b, c}, {a, b, T
f})
1 T2
Sim( T , T )
Jaccard co-efficient-based similarity function: 1 2
T1 T2
{c{c,
} d, e} 1
Ex. LetSim
T1 (=T {a,
b, c}, T2 =
1, T 2 )
0.2
{a, b, c, d , e}
94
Link Measure in ROCK
Clusters
C1:<a, b, c, d, e>: {a, b, c}, {a, b, d}, {a, b, e}, {a, c, d}, {a, c, e},
{a, d, e}, {b, c, d}, {b, c, e}, {b, d, e}, {c, d, e}
C2: <a, b, f, g>: {a, b, f}, {a, b, g}, {a, f, g}, {b, f, g}
Neighbors
Two transactions are neighbors if sim(T 1,T2) > threshold

Let T1 = {a, b, c}, T2 = {c, d, e}, T3 = {a, b, f}
T
1 connected to: {a,b,d}, {a,b,e}, {a,c,d}, {a,c,e}, {b,c,d},
{b,c,e}, {a,b,f}, {a,b,g}
T connected to: {a,c,d}, {a,c,e}, {a,d,e}, {b,c,e}, {b,d,e},
2
{b,c,d}
T connected to: {a,b,c}, {a,b,d}, {a,b,e}, {a,b,g}, {a,f,g}, {b,f,g}
3
Link Similarity
Link similarity between two transactions is the # of common

neighbors
link(T1, T2) = 4, since they have 4 common neighbors
{a, c, d}, {a, c, e}, {b, c, d}, {b, c, e}
link(T1, T3) = 3, since they have 3 common neighbors
{a, b, d}, {a, b, e}, {a, b, g}

95
Measuring Clustering Quality: External

Methods
Clustering quality measure: Q(C, Cg), for a clustering C

given the ground truth Cg.
Q is good if it satisfies the following 4 essential criteria

Cluster homogeneity: the purer, the better
Cluster completeness: should assign objects belong
to the same category in the ground truth to the
same cluster
Rag bag: putting a heterogeneous object into a pure
cluster should be penalized more than putting it into
a rag bag (i.e., miscellaneous or other category)
Small cluster preservation: splitting a small category
into pieces is more harmful than splitting a large
category into pieces
97
Assessing Clustering Tendency
Assess if non-random structure exists in the data by measuring the

probability that the data is generated by a uniform data
distribution
Test spatial randomness by statistic test: Hopkins Static
Given a dataset D regarded as a sample of a random variable o,
determine how far away o is from being uniformly distributed in
the data space
Sample n points, p , , p , uniformly from D. For each p , find
1
n
i
its nearest neighbor in D: xi = min{dist (pi, v)} where v in D
Sample n points, q1, , qn, uniformly from D. For each qi, find
its nearest neighbor in D {qi}: yi = min{dist (qi, v)} where v
in D and v qi
Calculate the Hopkins Statistic:
If D is uniformly distributed, xi and yi will be close to each
other and H is close to 0.5. If D is clustered, H is close to 1
98

Concepts and Techniques: - Chapter 10

Uploaded by

Copyright:

Available Formats

Concepts and Techniques: - Chapter 10

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Concepts and Techniques: - Chapter 10

Uploaded by

Copyright:

Available Formats

Data Mining:

Chapter 10. Cluster Analysis: Basic

Cluster Analysis: Basic Concepts

What is Cluster Analysis?

Cluster: A collection of data objects

Applications of Cluster Analysis

Summarization: Preprocessing for regression, PCA,

Compression: Image processing: vector quantization

Finding K-nearest Neighbors

Localizing search to one or a small number of clusters

Outlier detection: Outliers are often viewed as those far

Clustering: Application Examples

Biology: taxonomy of living things: kingdom, phylum, class,

Basic Steps to Develop a Clustering

Select info concerning the task of interest

Minimal information redundancy

Validation of the results

Expressed via a cost function or some rules

Similarity of two feature vectors

Validation test (also, clustering tendency test)

Interpretation of the results

Integration with applications

Quality: What Is Good Clustering?

A good clustering method will produce high quality

high intra-class similarity: cohesive within clusters

low inter-class similarity: distinctive between

The quality of a clustering method depends on

the similarity measure used by the method

its implementation, and

Its ability to discover some or all of the hidden

Measure the Quality of Clustering

Similarity is expressed in terms of a distance

The definitions of distance functions are usually

Weights should be associated with different

There is usually a separate quality function that

It is hard to define similar enough or good

The answer is typically highly subjective

Considerations for Cluster

Exclusive (e.g., one customer belongs to only one region)

Single level vs. hierarchical partitioning (often, multi-level

Distance-based (e.g., Euclidian, road network, vector) vs.

Full space (often when low dimensional) vs. subspaces

Requirements and Challenges

User may give inputs on constraints

Use domain knowledge to determine input parameters

Major Clustering Approaches (I)

Major Clustering Approaches (II)

Chapter 10. Cluster Analysis: Basic

Cluster Analysis: Basic Concepts

Partitioning Algorithms: Basic

Partitioning method: Partitioning a database D of n objects into

Given k, find a partition of k clusters that optimizes the chosen

Global optimal: exhaustively enumerate all partitions

Heuristic methods: k-means and k-medoids algorithms

k-means (MacQueen67, Lloyd57/82): Each cluster is

k-medoids or PAM (Partition around medoids) (Kaufman &

The K-Means Clustering

Given k, the k-means algorithm is implemented in