C4.5: programs for machine learning
C4.5: programs for machine learning
Deterministic part-of-speech tagging with finite-state transducers
Computational Linguistics
IGTree: Using Trees for Compression and Classification in Lazy LearningAlgorithms
Artificial Intelligence Review - Special issue on lazy learning
| Hi-index | 0.00 |
This paper describes the use of rule induction techniques for the automatic extraction of phonemic knowledge and rules from pairs of pronunciation lexica. This extracted knowledge allows the adaptation of speech processing systems to regional variants of a language. As a case study, we apply the approach to Northern Dutch and Flemish (the variant of Dutch spoken in Flanders, a part of Belgium), based on Celex and Fonilex, pronunciation lexica for Northern Dutch and Flemish, respectively. In our study, we compare two rule induction techniques, Transformation-Based Error-Driven Learning (TBEDL) (Brill, 1995) and C5.0 (Quinlan, 1993), and evaluate the extracted knowledge quantitatively (accuracy) and qualitatively (linguistic relevance of the rules). We conclude that, whereas classification-based rule induction with C5.0 is more accurate, the transformation rules learned with TBEDL can be more easily interpreted.