Clustering%20of%20Trajectory%20Data%20obtained%20from%20Soccer%20Game%20Record%20-A%20First%20Step%20to%20Behavioral%20Modeling

1
Clustering of Trajectory Data obtained from
Soccer Game Record -A First Step to Behavioral
Modeling                      
Shoji Hirano Shusaku Tsumotohirano_at_ieee.org
tsumoto_at_computer.org Dept of Medical
Informatics, Shimane Univ. School of Medicine,
Japan
2
Outline

• Introduction
• Data Structure
• Method
• Experimental Results
• Conclusions and Future Work

3
Introduction

• Clustering of Spatio-temporal Data
• Provides a way to discover interesting
  characteristics about the motion of targets
• Related field meteorology, medical image
  analysis, sports, crime research etc.
• Approaches
• Spatial clustering temporal continuity trace
  (e.g. tracking of moving object)
• Spatial clustering based on temporal correlation
  (e.g. fMRI analysis)
• Spatial clustering observation of the temporal
  changes of the clusters (e.g. Observation of
  the climate regimes)

4
Objective

• Development of a clustering method for
  trajectories with multiscale structural
  comparison scheme
• Compare trajectories according to both local and
  global views.
• Visualize common characteristics of trajectories
• Application Clustering of trajectories of passes
  in soccer game records
• Discovery of interesting spatio-temporal patterns
  of passes which may reflect the strategy and
  tactics of the team
• Globally similar passes strategy of the team
  -ex. Attack from right side
• Locally similar passes tactics of the ream -x.
  Frequent use of one-two passes

5
Data Structure

• Soccer game records(provided for research
  purpose by DataStadium Inc., Japan)

6
Data Structure

• Field geometry and Pass sequence

5346
Y
IN GOAL
PASS start
X
t
-3500
3500
-5346
7
Pass sequence clustering Problems

• Irregularly-sampled spatio-temporal sequence
• Data point is generated when a player takes an
  interaction with a ball
• High interaction -gt Dense DataLow interaction -gt
  Sparse Data
• Need for Multiscale Observation
• Strategy -gt global pass featureTactics -gt local
  pass feature
• Both exist concurrently
• It is required to partly change comparison scale
  according to the granularity of data and type of
  events

Dense
Sparse
8
Trajectory Mining
9
Method Multiscale Matching

• A pattern matching method that compares
  structural similarity of planar curves across
  multiple observation scales
• Able to compare objects by partly changing
  observation scales
• Simultaneously compare both global and local
  similarities

Scale s
Sequence A
Sequence B
10
Multiscale Description (Witkin et al 1984,
Mokhatan et al. 1986)

• Describe convex/concave structure at multiple
  scales
• Sequence description
• t course parameter
• Sequence x(t) at scale s
• Scale s controls the degree of smoothing
• s small local feature, s large global
  feature

Scale s
11
Multiscale Matching based on Convex/Concave
Structure of Segments (Ueda et al. 1990)

• Segment Partial sequence between adjacent
  inflection points
• Curvature K (t, s) at scale s
• Inflection point
• Represent a sequence as a set of segments

Scale s
12
Matching Procedure
13
Segment Dissimilarity

• Dissimilarity of Segments
• Dissimilarity of sequences

Max( , )
Rotation Angle
Length
P the number of matched pairs
14
Indiscernibility-based Clustering Overview

• Assignment of initial equivalence relations (ERs)
  
• Assign an initial ER to each of the N objects.
• An ER independently performs binary
  classification, similar or dissimilar, based on
  the relative proximity.
• Indiscernible objects under all of the N ERs form
  a cluster.

15
Experiments

• Data
• Game records of FIFA WorldCup 2002 (64 games,
  including all heats and finals)
• Number of goals 168 (own goals excluded)
• Procedure
• Select series containing IN GOAL event, and
  generate a total of 168 trajectories of 2-D ball
  location.
• For every possible pair of the trajectories,
  calculate dissimilarity by using multiscale
  matching.
• Group the trajectories by using the obtained
  dissimilarities and indiscernibility-based
  clustering

16
Experimental Results

• Cluster Constitution

Cluster Cases
1 87
2 24
3 17
4 16
5 8
6 4
Cluster Cases
7 3
8 3
9 2
10 2
11 2
12 1
Note 55.2 (7839/14196) of triplet in the
dissimilarity matrix did not satisfy
the triangular inequality due to matching failure
17
Experimental Results (contd)

• Cluster 1 (87 cases)

Corner Kick Goal
Matching Result
IN GOAL
Europe 45, South America 24, Asia 9
18
Experimental Results (contd)

• Cluster 2 (24 cases)

Complex Pass Side attack- Goal
Matching Result
IN GOAL
Germany vs Cameroon
Poland vs Portugal
Europe 13, South America 7, Asia 3
19
Experimental Results (contd)

• Cluster 4 (16 cases)

Side Change Centering/Dribble Goal
Matching Result
IN GOAL
Slovenia vs Paraguay
China vs Turkey
Europe 10, South America 4, Africa 2
20
Experimental Results (contd)

• Cluster 3 (17 cases)

Side Change Centering/Dribble
Goal (Intermediate cases between Cluster 2 and
4)
Europe 10, South America 2, Africa 2 Asia 2
21
Summary of Experimental Results

• Goal success patterns can be classified into 4
  major groups (with 8 minor patterns)
• Patterns complexity of pass sequences
• With additional information
• Dribble/Centering/Side change European Style
• However, the differences are not statistically
  significant.
• Key is Side Change
• Players (Defenders) should take care of the other
  side of the ball movement.
• The higher complexity of pass transactions, the
  higher rate of goal success gains by side change.

22
Conclusions

• Presented a new scheme of spatio-temporal data
  mining
• Grouped similar patterns using multiscale
  comparison and indiscernibility-based clustering
  techniques.
• Visualized similar patterns using matching
  results.
• Application to real World Cup data
• Grouping and visualization of interesting pass
  patternsex. Complex pass -gt side attack -gt goal

23
Future Work

• Technical Issues
• Numerical Evaluation
• Validation and improvement of segment
  dissimilarity measure inclusion of event type to
  dissimilarity
• Apply the proposed method to all path series
  including non-IN GOAL series
• Differences between success and failure are very
  small.
• This suggests that the patterns of soccer attack
  are simple.
• Apply the proposed method to medical environment
• Trajectories of Laboratory Examinations (IEEE
  ICDM06)
• Trajectories of Patients Movement Patient
  Safety

24
Matching Criteria

• Criteria for determining the best set of segment
  pairs
• Complete match original sequence should be
  correctly formed by concatenating the selected
  segments without any overlaps or gaps
• Minimization of total segment difference

Overlap
Gap
a2
a1
a4
a3
A
a5
b2
b4
P Number of matched segment pairs
b1
b3
b5
B
dissimiarity of segments
25
Matching Failure Problem in MSM

• Theoretically, any sequence can finally become a
  single segment at enough high scales. Therefore,
  any pair of sequences should be successfully
  matched.
• Practically, there should be an upper limit of
  scales in order to reduce computational
  complexity. Therefore, the number of segments can
  be different even at the highest scales.
• If matching is not successful, the method should
  return infinite dissimilarity or a magic value
  that indicates matching failure.

match
Scale n
Scale 2
no-match
Scale 1
26
Trajectory Mining