Efficient modeling of analogy

  • Authors:

  • Lars G. Johnsen;Christer Johansson

  • Affiliations:

  • Dept. of Linguistics and Literature, University of Bergen, Bergen, Norway;Dept. of Linguistics and Literature, University of Bergen, Bergen, Norway

  • Venue:
  • CICLing'05 Proceedings of the 6th international conference on Computational Linguistics and Intelligent Text Processing
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Analogical modeling (AM) is a memory based model with a documented performance comparable to other types of memory based learning. Known algorithms implementing AM have a computationally complexity of O(2n). We formulate a representation theorem on analogical modeling which is used for implementing a range of approximations to AM with a complexity starting as low as O(n).