Optimizing Constraint-Based Mining by Automatically Relaxing Constraints

In constraint-based mining, the monotone and antimonotone properties are exploited to reduce the search space. Even if a constraint has not such suitable properties, existing algorithms can be re-used thanks to an approximation, called relaxation. In this paper, we automatically compute monotone relaxations of primitive-based constraints. First, we show that the latter are a superclass of combinations of both kinds of monotone constraints. Second, we add two operators to detect the properties of monotonicity of such constraints. Finally, we define relaxing operators to obtain monotone relaxations of them.

• Type: Conference
• Year: 2005
• Where: ICDM
• Authors: Arnaud Soulet, Bruno Crémilleux
• Keywords: Constraint-based Mining, Data Mining, ICDM 2005, Monotone Constraints, Monotone Relaxations