Class imbalance problem in UCS classifier system: Fitness adaptation

The class imbalance problem has been said to challenge the performance of concept learning systems. Learning systems tend to be biased towards the majority class, and thus have poor generalization for the minority class instances. We analyze the class imbalance problem in learning classifier systems based on genetic algorithms. In particular we study UCS, a rule-based classifier system which learns under a supervised learning scheme. We analyze UCS on an artificial domain with varying imbalance levels. We find UCS fairly sensitive to high levels of class imbalance, to the degree that UCS tends to evolve a simple model of the feature space classified according to the majority class. We analyze strategies for dealing with class imbalances, and find fitness adaptation based on class-sensitive accuracy a useful tool for alleviating the effects of class imbalances.