Mining Financial Data with Scheduled Discovery of Exception Rules

This paper shows preliminary results, on financial data, of an algorithm for discovering pairs of an exception rule and a common sense rule under a prespecified schedule. An exception rule, which represents a regularity of exceptions to a common sense rule, often exhibits interestingness. Discovery of pairs of an exception rule and a common sense rule under threshold scheduling has been successful in efficient discovery of interesting rules. In this paper, we apply it to financial data, which has been provided as a benchmark data set for data mining methods. Examples of discovered knowledge as well as a simple description of the approach are both provided in this paper.