Automatically extracting and representing collocations for language generation
ACL '90 Proceedings of the 28th annual meeting on Association for Computational Linguistics
| Hi-index | 0.00 |
A major challenge in computational linguistics is to uncover word interactions in linguistic expressions. In this paper a new framework for discovering interaction between the words constituting multi-word relevant expressions is proposed. This framework is built on an algorithm for relevant expression extraction called LocalMaxs algorithm, partitioning round medoids clustering method and Bayesian networks. Bayesian networks are attractive for their ability to represent dependencies and to learn from observations. This new technology facilitates text comprehension. It may also enable control of highly ambiguous text input.