摘要 |
Embodiments of the invention disclose a system and a method for transforming a set of particles in an output set of particles representing a set of words suitable for use in an information retrieval system. The method generates, for each particle in the set of particles, combinations of parts of a particle, and replaces the particle in the set of particles with the parts of a combination maximizing a total minimum edit distance (MED) of the set of particles. For example, the method determines a MED of each particle in the set of particles, determines the total MED of the set of particles as summations of the MED of each particle, and then determines the combination maximizing the total MED of the set of particles.
|