A particle swarm based hybrid system for imbalanced medical data sampling

1
A particle swarm based hybrid system for
imbalanced medical data sampling

• Pengyi Yang
• yangpy_at_it.usyd.edu.au

School of Information Technologies The University
of Sydney, NSW 2006, Australia
NICTA, Australian Technology Park, Eveleigh NSW
2015, Australia
2
Road map

• Imbalanced class distribution in medical data
• Sampling
• Over-sampling
• Under-sampling
• Convert feature selection techniques as sampling
  strategy
• System overview
• Results
• Conclusion

3
Imbalanced class distribution in medical data

• Medical data are commonly with imbalanced class
  distribution
• Why?
• Positive samples are special cases (rare) while
  negative samples are abundant
• (Reversely, only positive samples are collected)
• Data contain subtypes each with limited samples

4
Problem

• Building classification model with imbalanced
  dataset will cause the under represented class
  been overlooked or even ignored.
• Yet, the rare classes often carry important
  biological implication.
• The difficulty becomes how to remedy the
  imbalanced class distribution.

5
Remedy

• Via sampling before model building process
• Over sampling increase sample size of minority
  class (could introduce noise and redundancy)
• Under sampling decrease sample size of majority
  class (could remove representative samples)
• Via cost-sensitive learning within model
  building process
• Need to choose an appropriate cost-metric (hard
  to determine a prior)

6
Current methods

• The most straightforward way random
  over-sampling and under-sampling
• Naive method but work well in different
  situations
• Clustering and sampling
• Clustering dataset and sampling according to the
  characteristic of each cluster
• Synthesizing new examples
• Most popular is smote which creates
  artificial samples to increase the size of
  minority class

7
Our contribution Proposing a novel sampling
strategy

• Convert feature selection technique as sampling
  strategy
• Selecting a subset of optimal samples from
  majority class

Supervised sample selection
(balanced dataset)
(imbalanced dataset)
8
Conceptual representation
9
Particle swarm optimization

• Problem encoding
• Each particle is a subset of samples from the
  majority class

m is the sample size of majority class
10
Final Schema
11
Results
(1) PSO achieved better classification
results.
(2) Different evaluation metrics could
gives a different evaluation
indication.
12
Results continue
(3) Different classifiers also perform
differently within the same sampling method
13
Key observation

• The evaluation of data sampling strategy is
  compounded by the type of classifier applied and
  the evaluation metric used.
• Therefore, caution should be drawn when the
  conclusion is made on the basis of a single type
  of classifier or evaluation metric.

14
Conclusion

• The study shows that with proper modification
  feature selection techniques can be applied to
  sampling of imbalanced data.
• The application of such technique to medical
  domain demonstrates it can help to increase the
  classification accuracy which is valuable to
  prediction or decision support systems.

15
Publication

• P Yang, L Hsu, B. Zhou, Z. Zhang, A Zomaya, A
  particle swarm based hybrid system for imbalanced
  medical data sampling, accepted by BMC Genomics.

16
Questions!