Discovering Knowledge from Local Patterns with Global Constraints

Abstract

It is well known that local patterns are at the core of a lot of
  knowledge which may be discovered from data. Nevertheless, use of local
  patterns is limited by 
  their huge number and computational costs.  Several approaches (e.g.,
  condensed representations, pattern set discovery) aim at grouping or
  synthesizing local patterns to provide a global view of the data.  A
  global pattern is a pattern which is a set or a synthesis of local
  patterns coming from the data.  In this paper, we propose the idea of
  global constraints to write queries addressing global patterns.  A key
  point is the ability to bias the designing of global patterns according
  to the expectation of the user.  For instance, a global pattern can be
  oriented towards the search of exceptions or a clustering.  It requires
  to write queries taking into account such biases. Open issues are to
  design a generic framework to express powerful global constraints and
  solvers to mine them.  We think that global constraints are a promising
  way to discover relevant global patterns.