Data Mining Cluster Analysis: Basic Concepts and Algorithms Lecture Notes for Chapter 8 Introduction to Data Mining
by Tan, Steinbach, Kumar
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
What is Cluster Analysis?
Finding groups of objects such that the objects in a group will be similar (or related) to one another and different from (or unrelated to) the objects in other groups
Intra-cluster distances are minimized Inter-cluster distances are maximized
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Applications of Cluster Analysis
Understanding
Group related documents for browsing, group genes and proteins that have similar functionality, or group stocks with similar price fluctuations
Discovered Clusters
Industry Group
1 2 3 4
Applied-Matl-DOWN,Bay-Network-Down,3-COM-DOWN, Cabletron-Sys-DOWN,CISCO-DOWN,HP-DOWN, DSC-Comm-DOWN,INTEL-DOWN,LSI-Logic-DOWN, Micron-Tech-DOWN,Texas-Inst-Down,Tellabs-Inc-Down, Natl-Semiconduct-DOWN,Oracl-DOWN,SGI-DOWN, Sun-DOWN Apple-Comp-DOWN,Autodesk-DOWN,DEC-DOWN, ADV-Micro-Device-DOWN,Andrew-Corp-DOWN, Computer-Assoc-DOWN,Circuit-City-DOWN, Compaq-DOWN, EMC-Corp-DOWN, Gen-Inst-DOWN, Motorola-DOWN,Microsoft-DOWN,Scientific-Atl-DOWN Fannie-Mae-DOWN,Fed-Home-Loan-DOWN, MBNA-Corp-DOWN,Morgan-Stanley-DOWN Baker-Hughes-UP,Dresser-Inds-UP,Halliburton-HLD-UP, Louisiana-Land-UP,Phillips-Petro-UP,Unocal-UP, Schlumberger-UP
Technology1-DOWN
Technology2-DOWN
Financial-DOWN Oil-UP
Summarization
Reduce the size of large data sets
Clustering precipitation in Australia
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
What is not Cluster Analysis?
Supervised classification
Have class label information
Simple segmentation
Dividing students into different registration groups alphabetically, by last name
Results of a query
Groupings are a result of an external specification
Graph partitioning
Some mutual relevance and synergy, but areas are not identical
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Notion of a Cluster can be Ambiguous
How many clusters?
Six Clusters
Two Clusters
Four Clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Types of Clusterings
A clustering is a set of clusters Important distinction between hierarchical and partitional sets of clusters Partitional Clustering
A division data objects into non-overlapping subsets (clusters) such that each data object is in exactly one subset
Hierarchical clustering
A set of nested clusters organized as a hierarchical tree
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Partitional Clustering
Original Points
A Partitional Clustering
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering
p1 p3 p2 p4
p1 p2
Traditional Hierarchical Clustering
p3 p4
Traditional Dendrogram
p1 p3 p2 p4
p1 p2
Non-traditional Hierarchical Clustering
Tan,Steinbach, Kumar
p3 p4
Non-traditional Dendrogram
4/18/2004 #
Introduction to Data Mining
Other Distinctions Between Sets of Clusters
Exclusive versus non-exclusive
In non-exclusive clusterings, points may belong to multiple clusters. Can represent multiple classes or border points
Fuzzy versus non-fuzzy
In fuzzy clustering, a point belongs to every cluster with some weight between 0 and 1 Weights must sum to 1 Probabilistic clustering has similar characteristics
Partial versus complete
In some cases, we only want to cluster some of the data
Heterogeneous versus homogeneous
Cluster of widely different sizes, shapes, and densities
#
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Types of Clusters
Well-separated clusters Center-based clusters Contiguous clusters Density-based clusters Property or Conceptual Described by an Objective Function
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Types of Clusters: Well-Separated
Well-Separated Clusters:
A cluster is a set of points such that any point in a cluster is closer (or more similar) to every other point in the cluster than to any point not in the cluster.
3 well-separated clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Types of Clusters: Center-Based
Center-based
A cluster is a set of objects such that an object in a cluster is closer (more similar) to the center of a cluster, than to the center of any other cluster The center of a cluster is often a centroid, the average of all the points in the cluster, or a medoid, the most representative point of a cluster
4 center-based clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Types of Clusters: Contiguity-Based
Contiguous Cluster (Nearest neighbor or Transitive)
A cluster is a set of points such that a point in a cluster is closer (or more similar) to one or more other points in the cluster than to any point not in the cluster.
8 contiguous clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Types of Clusters: Density-Based
Density-based
A cluster is a dense region of points, which is separated by low-density regions, from other regions of high density.
Used when the clusters are irregular or intertwined, and when noise and outliers are present.
6 density-based clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Types of Clusters: Conceptual Clusters
Shared Property or Conceptual Clusters
Finds clusters that share some common property or represent a particular concept.
2 Overlapping Circles
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Types of Clusters: Objective Function
Clusters Defined by an Objective Function
Finds clusters that minimize or maximize an objective function. Enumerate all possible ways of dividing the points into clusters and evaluate the `goodness' of each potential set of clusters by using the given objective function. (NP Hard) Can have global or local objectives.
Hierarchical clustering algorithms typically have local objectives
Partitional algorithms typically have global objectives
A variation of the global objective function approach is to fit the data to a parameterized model.
Parameters for the model are determined from the data.
Mixture models assume that the data is a mixture' of a number of statistical distributions.
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Types of Clusters: Objective Function
Map the clustering problem to a different domain and solve a related problem in that domain
Proximity matrix defines a weighted graph, where the nodes are the points being clustered, and the weighted edges represent the proximities between points Clustering is equivalent to breaking the graph into connected components, one for each cluster.
Want to minimize the edge weight between clusters and maximize the edge weight within clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Characteristics of the Input Data Are Important
Type of proximity or density measure
This is a derived measure, but central to clustering
Sparseness
Dictates type of similarity Adds to efficiency
Attribute type
Dictates type of similarity
Type of Data
Dictates type of similarity Other characteristics, e.g., autocorrelation
Dimensionality Noise and Outliers Type of Distribution
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Clustering Algorithms
K-means and its variants Hierarchical clustering Density-based clustering
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
K-means Clustering
Partitional clustering approach Each cluster is associated with a centroid (center point) Each point is assigned to the cluster with the closest centroid Number of clusters, K, must be specified The basic algorithm is very simple
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
K-means Clustering Details
Initial centroids are often chosen randomly.
Clusters produced vary from one run to another.
The centroid is (typically) the mean of the points in the cluster. Closeness is measured by Euclidean distance, cosine similarity, correlation, etc. K-means will converge for common similarity measures mentioned above. Most of the convergence happens in the first few iterations.
Often the stopping condition is changed to Until relatively few points change clusters n = number of points, K = number of clusters, I = number of iterations, d = number of attributes
Introduction to Data Mining 4/18/2004 #
Complexity is O( n * K * I * d )
Tan,Steinbach, Kumar
Two different K-means Clusterings
3 2.5
Original Points
1.5
y
1 0.5 0 -2
-1.5
-1
-0.5
0.5
1.5
2.5
2.5
1.5
1.5
y
1 0.5 0.5 0 0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -2
-1.5
-1
-0.5
0.5
1.5
Optimal Clustering
Tan,Steinbach, Kumar Introduction to Data Mining
Sub-optimal Clustering
4/18/2004 #
Importance of Choosing Initial Centroids
Iteration 1 6 5 4 3 2
3 2.5
1.5
y
1 0.5 0 -2
-1.5
-1
-0.5
0.5
1.5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Importance of Choosing Initial Centroids
Iteration 1
3 3 2.5 2.5
Iteration 2
3 2.5
Iteration 3
1.5
1.5
1.5
y
1 0.5 0.5 0 0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -2
0.5
-2
-1.5
-1
-0.5
0.5
1.5
-1.5
-1
-0.5
0.5
1.5
Iteration 4
3 3 2.5 2.5
Iteration 5
3 2.5
Iteration 6
1.5
1.5
1.5
y
1 0.5 0.5 0 0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -2
0.5
-2
-1.5
-1
-0.5
0.5
1.5
-1.5
-1
-0.5
0.5
1.5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Evaluating K-means Clusters
Most common measure is Sum of Squared Error (SSE)
For each point, the error is the distance to the nearest cluster To get SSE, we square these errors and sum them.
SSE dist 2 (mi , x)
i 1 xCi
x is a data point in cluster Ci and mi is the representative point for cluster Ci
can show that mi corresponds to the center (mean) of the cluster
Given two clusters, we can choose the one with the smallest error One easy way to reduce SSE is to increase K, the number of clusters
A good clustering with smaller K can have a lower SSE than a poor clustering with higher K
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Importance of Choosing Initial Centroids
Iteration 1 5 4 3 2
3 2.5
1.5
y
1 0.5 0 -2
-1.5
-1
-0.5
0.5
1.5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Importance of Choosing Initial Centroids
Iteration 1
3 3 2.5 2.5
Iteration 2
1.5
1.5
y
1 0.5 0.5 0 0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -2
-1.5
-1
-0.5
0.5
1.5
Iteration 3
3 3 2.5 2.5
Iteration 4
3 2.5
Iteration 5
1.5
1.5
1.5
y
1 0.5 0.5 0 0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -2
0.5
-2
-1.5
-1
-0.5
0.5
1.5
-1.5
-1
-0.5
0.5
1.5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Problems with Selecting Initial Points
If there are K real clusters then the chance of selecting one centroid from each cluster is small.
Chance is relatively small when K is large
If clusters are the same size, n, then
For example, if K = 10, then probability = 10!/1010 = 0.00036 Sometimes the initial centroids will readjust themselves in right way, and sometimes they dont
Consider an example of five pairs of clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
10 Clusters Example
Iteration 1 4 3 2
8 6 4 2
0 -2 -4 -6
10
15
20
x Starting with two initial centroids in one cluster of each pair of clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
10 Clusters Example
Iteration 1
8 6 4 2 8 6 4 2
Iteration 2
0 -2 -4 -6
y
0 5 10 15 20
0 -2 -4 -6
10
15
20
Iteration 3
8 6 4 2 8 6 4 2
Iteration 4
0 -2 -4 -6
y
0 5 10 15 20
0 -2 -4 -6
10
15
20
Starting with two initial centroids in one cluster of each pair of clusters
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
10 Clusters Example
Iteration 1 4 3 2
8 6 4 2
0 -2 -4 -6
10
15
20
x
Starting with some pairs of clusters having three initial centroids, while other have only one.
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
10 Clusters Example
Iteration 1
8 6 4 2 8 6 4 2
Iteration 2
0 -2 -4 -6
y
0 5 10 15 20
0 -2 -4 -6
0 8 6 4 2
10
15
20
x Iteration 3
8 6 4 2
x Iteration 4
0 -2 -4 -6
y
0 5 10 15 20
0 -2 -4 -6
10
15
20
Starting with some pairs of clusters having three initial centroids, while other have only one.
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Solutions to Initial Centroids Problem
Multiple runs
Helps, but probability is not on your side
Sample and use hierarchical clustering to determine initial centroids Select more than k initial centroids and then select among these initial centroids
Select most widely separated
Postprocessing Bisecting K-means
Not as susceptible to initialization issues
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Handling Empty Clusters
Basic K-means algorithm can yield empty clusters Several strategies
Choose the point that contributes most to SSE Choose a point from the cluster with the highest SSE If there are several empty clusters, the above can be repeated several times.
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Updating Centers Incrementally
In the basic K-means algorithm, centroids are updated after all points are assigned to a centroid An alternative is to update the centroids after each assignment (incremental approach)
Each assignment updates zero or two centroids More expensive Introduces an order dependency Never get an empty cluster Can use weights to change the impact
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Pre-processing and Post-processing
Pre-processing
Normalize the data Eliminate outliers
Post-processing
Eliminate small clusters that may represent outliers Split loose clusters, i.e., clusters with relatively high SSE Merge clusters that are close and that have relatively low SSE Can use these steps during the clustering process
ISODATA
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Bisecting K-means
Bisecting K-means algorithm
Variant of K-means that can produce a partitional or a hierarchical clustering
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Bisecting K-means Example
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of K-means
K-means has problems when clusters are of differing
Sizes Densities Non-globular shapes
K-means has problems when the data contains outliers.
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of K-means: Differing Sizes
Original Points
K-means (3 Clusters)
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of K-means: Differing Density
Original Points
K-means (3 Clusters)
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of K-means: Non-globular Shapes
Original Points
K-means (2 Clusters)
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Overcoming K-means Limitations
Original Points
K-means Clusters
One solution is to use many clusters. Find parts of clusters, but need to put together.
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Overcoming K-means Limitations
Original Points
K-means Clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Overcoming K-means Limitations
Original Points
K-means Clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering
Produces a set of nested clusters organized as a hierarchical tree Can be visualized as a dendrogram
A tree like diagram that records the sequences of merges or splits
6
0.2
4 3 2 5 2 1 3 1 4
0.15
0.1
0.05
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Strengths of Hierarchical Clustering
Do not have to assume any particular number of clusters
Any desired number of clusters can be obtained by cutting the dendogram at the proper level
They may correspond to meaningful taxonomies
Example in biological sciences (e.g., animal kingdom, phylogeny reconstruction, )
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering
Two main types of hierarchical clustering
Agglomerative:
Start with the points as individual clusters
At each step, merge the closest pair of clusters until only one cluster (or k clusters) left
Divisive:
Start with one, all-inclusive cluster
At each step, split a cluster until each cluster contains a point (or there are k clusters)
Traditional hierarchical algorithms use a similarity or distance matrix
Merge or split one cluster at a time
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Agglomerative Clustering Algorithm
More popular hierarchical clustering technique Basic algorithm is straightforward
1. 2. 3. 4. Compute the proximity matrix Let each data point be a cluster Repeat Merge the two closest clusters
5.
6.
Update the proximity matrix
Until only a single cluster remains
Key operation is the computation of the proximity of two clusters
Different approaches to defining the distance between clusters distinguish the different algorithms
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Starting Situation
Start with clusters of individual points and a proximity matrix p1 p2 p3 p4 p5
p1
...
p2
p3 p4 p5
. . .
Proximity Matrix
...
p1 p2 p3 p4 p9 p10 p11 p12
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Intermediate Situation
After some merging steps, we have some clusters
C1 C1 C2 C3 C3 C4 C4 C5 C2 C3 C4 C5
Proximity Matrix
C1
C2
C5
...
p1 p2 p3 p4 p9 p10 p11 p12
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Intermediate Situation
We want to merge the two closest clusters (C2 and C5) and update the proximity matrix. C1 C2 C3 C4 C5
C1 C2 C3 C3 C4
C4
C5
Proximity Matrix
C1
C2
C5
...
p1 p2 p3 p4 p9 p10 p11 p12
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
After Merging
The question is How do we update the proximity matrix?
C1 C1 C3 C4 C2 U C5 C3 C4 C1 ? C2 U C5 ? ? ? ? ? ? C3 C4
Proximity Matrix
C2 U C5
...
p1 p2 p3 p4 p9 p10 p11 p12
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
How to Define Inter-Cluster Similarity
p1 p2 p3 p4 p5
...
Similarity?
p1 p2 p3 p4
MIN . MAX . Group Average . Proximity Matrix Distance Between Centroids Other methods driven by an objective function
Wards Method uses squared error
p5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
How to Define Inter-Cluster Similarity
p1 p1 p2 p3 p4 p2 p3 p4 p5
...
MIN . MAX . Group Average . Proximity Matrix Distance Between Centroids Other methods driven by an objective function
Wards Method uses squared error
p5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
How to Define Inter-Cluster Similarity
p1 p1 p2 p3 p4 p2 p3 p4 p5
...
MIN . MAX . Group Average . Proximity Matrix Distance Between Centroids Other methods driven by an objective function
Wards Method uses squared error
p5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
How to Define Inter-Cluster Similarity
p1 p1 p2 p3 p4 p2 p3 p4 p5
...
MIN . MAX . Group Average . Proximity Matrix Distance Between Centroids Other methods driven by an objective function
Wards Method uses squared error
p5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
How to Define Inter-Cluster Similarity
p1 p1 p2 p3 p4 p2 p3 p4 p5
...
MIN . MAX . Group Average . Proximity Matrix Distance Between Centroids Other methods driven by an objective function
Wards Method uses squared error
p5
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Cluster Similarity: MIN or Single Link
Similarity of two clusters is based on the two most similar (closest) points in the different clusters
Determined by one pair of points, i.e., by one link in the proximity graph.
I1 1.00 0.90 0.10 0.65 0.20 I2 0.90 1.00 0.70 0.60 0.50 I3 0.10 0.70 1.00 0.40 0.30 I4 0.65 0.60 0.40 1.00 0.80 I5 0.20 0.50 0.30 0.80 1.00
I1 I2 I3 I4 I5
4
4/18/2004
5
#
Tan,Steinbach, Kumar
Introduction to Data Mining
Hierarchical Clustering: MIN
5
3
5 2 4 4
0 3 6 2 5 4 1 0.2
2 3
1 6
0.15
0.1
0.05
Nested Clusters
Tan,Steinbach, Kumar Introduction to Data Mining
Dendrogram
4/18/2004 #
Strength of MIN
Original Points
Two Clusters
Can handle non-elliptical shapes
#
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of MIN
Original Points
Two Clusters
Sensitive to noise and outliers
#
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Cluster Similarity: MAX or Complete Linkage
Similarity of two clusters is based on the two least similar (most distant) points in the different clusters
Determined by all pairs of points in the two clusters
I1 I2 I3 I4 I5 I1 1.00 0.90 0.10 0.65 0.20 I2 0.90 1.00 0.70 0.60 0.50 I3 0.10 0.70 1.00 0.40 0.30 I4 0.65 0.60 0.40 1.00 0.80 I5 0.20 0.50 0.30 0.80 1.00
Tan,Steinbach, Kumar Introduction to Data Mining
5
#
4/18/2004
Hierarchical Clustering: MAX
4 2 5 2 3 3 4
1 5
0.4 0.35 0.3 0.25
0.2 0.15 0.1 0.05 0 3 6 4 1 2 5
Nested Clusters
Dendrogram
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Strength of MAX
Original Points
Two Clusters
Less susceptible to noise and outliers
#
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Limitations of MAX
Original Points Tends to break large clusters Biased towards globular clusters
Tan,Steinbach, Kumar Introduction to Data Mining
Two Clusters
4/18/2004
Cluster Similarity: Group Average
Proximity of two clusters is the average of pairwise proximity between points in the two clusters.
proximity(Clusteri , Clusterj )
piClusteri p jCluster j
proximity(p , p )
i j
|Clusteri ||Clusterj |
Need to use average connectivity for scalability since total proximity favors large clusters
I1 I2 I3 I4 I5
I1 1.00 0.90 0.10 0.65 0.20
I2 0.90 1.00 0.70 0.60 0.50
I3 0.10 0.70 1.00 0.40 0.30
I4 0.65 0.60 0.40 1.00 0.80
I5 0.20 0.50 0.30 0.80 1.00
4
4/18/2004
5
#
Tan,Steinbach, Kumar
Introduction to Data Mining
Hierarchical Clustering: Group Average
5 2
5 2
1
0.25 0.2 0.15
3 1
0.1 0.05 0
4
3
Nested Clusters
Dendrogram
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering: Group Average
Compromise between Single and Complete Link Strengths
Less susceptible to noise and outliers
Limitations
Biased towards globular clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering: Time and Space requirements
O(N2) space since it uses the proximity matrix.
N is the number of points.
O(N3) time in many cases
There are N steps and at each step the size, N2, proximity matrix must be updated and searched Complexity can be reduced to O(N2 log(N) ) time for some approaches
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Hierarchical Clustering: Problems and Limitations
Once a decision is made to combine two clusters, it cannot be undone No objective function is directly minimized Different schemes have problems with one or more of the following:
Sensitivity to noise and outliers Difficulty handling different sized clusters and convex shapes Breaking large clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
MST: Divisive Hierarchical Clustering
Build MST (Minimum Spanning Tree)
Start with a tree that consists of any point In successive steps, look for the closest pair of points (p, q) such that one point (p) is in the current tree but the other (q) is not Add q to the tree and put an edge between p and q
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
MST: Divisive Hierarchical Clustering
Use MST for constructing hierarchy of clusters
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
DBSCAN
DBSCAN is a density-based algorithm.
Density = number of points within a specified radius (Eps)
A point is a core point if it has more than a specified number of points (MinPts) within Eps
These are points that are at the interior of a cluster
A border point has fewer than MinPts within Eps, but is in the neighborhood of a core point
A noise point is any point that is not a core point or a border point.
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
DBSCAN: Core, Border, and Noise Points
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
DBSCAN Algorithm
Eliminate noise points Perform clustering on the remaining points
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
DBSCAN: Core, Border and Noise Points
Original Points
Point types: core, border and noise Eps = 10, MinPts = 4
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
When DBSCAN Works Well
Original Points
Clusters
Resistant to Noise
Can handle clusters of different shapes and sizes
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
DBSCAN: Determining EPS and MinPts
Idea is that for points in a cluster, their kth nearest neighbors are at roughly the same distance Noise points have the kth nearest neighbor at farther distance So, plot sorted distance of every point to its kth nearest neighbor
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Cluster Validity
For supervised classification we have a variety of measures to evaluate how good our model is
Accuracy, precision, recall
For cluster analysis, the analogous question is how to evaluate the goodness of the resulting clusters? But clusters are in the eye of the beholder! Then why do we want to evaluate them?
To avoid finding patterns in noise To compare clustering algorithms To compare two sets of clusters To compare two clusters
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Clusters found in Random Data
1 0.9 0.8 0.7
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
0 0.2 0.4 0.6 0.8 1
Random Points
0.6 0.5 0.4 0.3 0.2 0.1 0
DBSCAN
0.2
0.4
0.6
0.8
x
1 0.9 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.2 0.4 0.6 0.8 1 0 0 0.2 0.4
K-means
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Complete Link
0.6
0.8
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Different Aspects of Cluster Validation
1. Determining the clustering tendency of a set of data, i.e., distinguishing whether non-random structure actually exists in the data. Comparing the results of a cluster analysis to externally known results, e.g., to externally given class labels. Evaluating how well the results of a cluster analysis fit the data without reference to external information.
- Use only the data
2. 3.
4.
5.
Comparing the results of two different sets of cluster analyses to determine which is better.
Determining the correct number of clusters.
For 2, 3, and 4, we can further distinguish whether we want to evaluate the entire clustering or just individual clusters.
Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 #
Measures of Cluster Validity
Numerical measures that are applied to judge various aspects of cluster validity, are classified into the following three types.
External Index: Used to measure the extent to which cluster labels match externally supplied class labels.
Entropy
Internal Index: Used to measure the goodness of a clustering structure without respect to external information.
Sum of Squared Error (SSE)
Relative Index: Used to compare two different clusterings or clusters.
Often an external or internal index is used for this function, e.g., SSE or entropy
Sometimes these are referred to as criteria instead of indices
However, sometimes criterion is the general strategy and index is the numerical measure that implements the criterion.
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Measuring Cluster Validity Via Correlation
Two matrices
Proximity Matrix Incidence Matrix
One row and one column for each data point An entry is 1 if the associated pair of points belong to the same cluster An entry is 0 if the associated pair of points belongs to different clusters
Compute the correlation between the two matrices
Since the matrices are symmetric, only the correlation between n(n-1) / 2 entries needs to be calculated.
High correlation indicates that points that belong to the same cluster are close to each other. Not a good measure for some density or contiguity based clusters.
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
Measuring Cluster Validity Via Correlation
Correlation of incidence and proximity matrices for the K-means clusterings of the following two data sets.
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1
Corr = 0.9235
Tan,Steinbach, Kumar Introduction to Data Mining
Corr = 0.5810
4/18/2004 #
Using Similarity Matrix for Cluster Validation
Order the similarity matrix with respect to cluster labels and inspect visually.
1 0.9 0.8 0.7 10 20 30 40 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 20 40 60 80 0 100 Similarity
Points
0 0.2 0.4 0.6 0.8 1
0.6 0.5 0.4 0.3 0.2 0.1 0
50 60 70 80 90 100
Points
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Using Similarity Matrix for Cluster Validation
Clusters in random data are not so crisp
1 10 20 30 40 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 20 40 60 80 0 100 Similarity 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1
Points
50 60 70 80 90 100
Points
DBSCAN
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Using Similarity Matrix for Cluster Validation
Clusters in random data are not so crisp
1 10 20 30 40 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 20 40 60 80 0 100 Similarity 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1
Points
50 60 70 80 90 100
Points
K-means
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Using Similarity Matrix for Cluster Validation
Clusters in random data are not so crisp
1 10 20 30 40 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 20 40 60 80 0 100 Similarity 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1
Points
50 60 70 80 90 100
Points
Complete Link
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Using Similarity Matrix for Cluster Validation
1 0.9
1 2 6
500
0.8 0.7 0.6
1000
3 4
1500
0.5 0.4
2000
0.3 0.2 0.1
5
2500
7
3000 500 1000 1500 2000 2500 3000 0
DBSCAN
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Internal Measures: SSE
Clusters in more complicated figures arent well separated
Internal Index: Used to measure the goodness of a clustering structure without respect to external information
SSE
SSE is good for comparing two clusterings or two clusters (average SSE). Can also be used to estimate the number of clusters
10
6 4 2
9 8 7 6
SSE
5 4
0 -2 -4 -6 5 10 15
3 2 1 0 2 5 10 15 20 25 30
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Internal Measures: SSE
SSE curve for a more complicated data set
1 2 6
3 4
SSE of clusters found using K-means
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Internal Measures: Cohesion and Separation
Cluster Cohesion: Measures how closely related are objects in a cluster
Example: SSE
Cluster Separation: Measure how distinct or wellseparated a cluster is from other clusters
Example: Squared Error
Cohesion is measured by the within cluster sum of squares (SSE)
WSS ( x mi )2
i xC i
Separation is measured by the between cluster sum of squares
BSS Ci ( m mi )
i
Tan,Steinbach, Kumar
Where |Ci| is the size of cluster i
Introduction to Data Mining 4/18/2004 #
Internal Measures: Cohesion and Separation
Example: SSE
BSS + WSS = constant
m 3 4
1 m1 2 K=1 cluster:
m2 5
WSS (1 3)2 (2 3)2 (4 3)2 (5 3)2 10 BSS 4 (3 3)2 0 Total 10 0 10
K=2 clusters:
WSS (1 1.5)2 (2 1.5)2 (4 4.5)2 (5 4.5)2 1 BSS 2 (3 1.5)2 2 (4.5 3)2 9 Total 1 9 10
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Internal Measures: Cohesion and Separation
A proximity graph based approach can also be used for cohesion and separation.
Cluster cohesion is the sum of the weight of all links within a cluster. Cluster separation is the sum of the weights between nodes in the cluster and nodes outside the cluster.
cohesion
separation
#
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Internal Measures: Silhouette Coefficient
Silhouette Coefficient combine ideas of both cohesion and separation, but for individual points, as well as clusters and clusterings For an individual point, i
Calculate a = average distance of i to the points in its cluster Calculate b = min (average distance of i to points in another cluster) The silhouette coefficient for a point is then given by s = 1 a/b if a < b,
(or s = b/a - 1 if a b, not the usual case)
b a
Typically between 0 and 1. The closer to 1 the better.
Can calculate the Average Silhouette width for a cluster or a clustering
Introduction to Data Mining 4/18/2004 #
Tan,Steinbach, Kumar
External Measures of Cluster Validity: Entropy and Purity
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
Final Comment on Cluster Validity
The validation of clustering structures is the most difficult and frustrating part of cluster analysis.
Without a strong effort in this direction, cluster analysis will remain a black art accessible only to those true believers who have experience and great courage.
Algorithms for Clustering Data, Jain and Dubes
Tan,Steinbach, Kumar
Introduction to Data Mining
4/18/2004
