Beruflich Dokumente
Kultur Dokumente
CIKM06
Bagnall 03, Mahoney 05, Rodrigues et al. 04 moved away from STS
November 8, 2006
CIKM06
STS clustering
algorithm
Clusters:
November 8, 2006
CIKM06
Outline
1.
2.
Introduction
New Distance Measure for Cluster Sets
based on the notion of cluster shapes
3.
4.
November 8, 2006
CIKM06
STS Clustering
November 8, 2006
CIKM06
K-means Clustering
Result: a set of K
cluster centers
Cluster Centers
November 8, 2006
CIKM06
November 8, 2006
CIKM06
cluster_set_dist(B,A)
X
A
Y
November 8, 2006
November 8, 2006
CIKM06
10
Cluster Structure
November 8, 2006
CIKM06
11
k=3 w=8
k=3 w=16
k=4 w=8
November 8, 2006
CIKM06
12
Outline
1.
2.
3.
4.
Introduction
New Distance Measure for Cluster Sets
STS Cluster Matching
Observations and Conclusions
November 8, 2006
CIKM06
13
Matching algorithm:
Outputs a guess -- which of the N time series in the
dataset produced the query?
Algorithm accuracy:
Percentage of times that the matching algorithm is correct.
Note: no previous work succeeded to attain high accuracy,
even with dataset of size 2!
November 8, 2006
CIKM06
14
Matching Algorithm
1.
2.
Pre-processing phase:
For each sequence in the dataset, perform Q
clustering runs with given K and w, and calculate its
cluster structure.
Store all the structures in a master table.
Matching phase:
1. Given a query, find the Euclidean distance from its
shape to each of the structures in the master table.
2. Return the sequence whose structure is the closest.
November 8, 2006
CIKM06
15
Example
Master table
k=3 w=8
November 8, 2006
CIKM06
16
Performance Evaluation
November 8, 2006
CIKM06
17
Outline
1.
2.
3.
4.
Introduction
New Distance Measure for Cluster Sets
STS Cluster Matching Algorithm
Observations and Conclusions
November 8, 2006
CIKM06
18
Conclusions
November 8, 2006
CIKM06
19
Future Work
WHY?
HOW?
November 8, 2006
20
Questions?
Thank you!