Projects per year
Abstract
Predicting overlapping trigrams of class labels is a recently-proposed method to improve performance on sequence labelling tasks. In this method, sequence elements are effectively classified three times, therefore some procedure is needed to post-process those overlapping classifications into one output sequence. In this paper, we present a rule-based procedure learned automatically from training data. In combination with a memory-based leaner predicting class trigrams, the performance of this meta-learned overlapping trigram post-processor matches that of a handcrafted post-processing rule used in the original study on class trigrams. Moreover, on two domain-specific entity chunking tasks, the class trigram method with automatically learned post-processing rules compares favourably with recent probabilistic sequence labelling techniques, such as maximum-entropy markov models and conditional random fields.
Original language | English |
---|---|
Title of host publication | Proceedings of the Fourth Learning Language in Logic Workshop |
Editors | J. Cussens, C. Nedellec |
Place of Publication | Bonn |
Publisher | [s.n.] |
Pages | 3-10 |
Number of pages | 8 |
Publication status | Published - 2005 |
Fingerprint
Dive into the research topics of 'Rule meta-learning for trigram-based sequence processing'. Together they form a unique fingerprint.Projects
- 3 Finished
-
Optimization in machine learning of language
Daelemans, W. M. P.
1/01/04 → 1/01/06
Project: Research project
-
Rolaquad: Robust language understanding for question - answering dialogues.
Canisius, S. V. M., Daelemans, W. M. P., Lendvai, P. K. & van den Bosch, A.
1/01/04 → 1/01/08
Project: Research project
-