4 papers · 1 filter
Forgetting Exceptions is Harmful in Language Learning
Walter Daelemans, Antal van den Bosch, Jakub Zavrel
We show that in language learning, contrary to received wisdom, keeping exceptional training instances in memory can be beneficial for generalization accuracy. We investigate this…
Improving Data Driven Wordclass Tagging by System Combination
Hans van Halteren, Jakub Zavrel, Walter Daelemans
In this paper we examine how the differences in modelling between different data driven systems performing the same NLP task can be exploited to yield a higher accuracy than the be…
Modularity in inductively-learned word pronunciation systems
Antal van den Bosch, Ton Weijters, Walter Daelemans
In leading morpho-phonological theories and state-of-the-art text-to-speech systems it is assumed that word pronunciation cannot be learned or performed without in-between analyses…
Do not forget: Full memory in memory-based learning of word pronunciation
Antal van den Bosch, Walter Daelemans
Memory-based learning, keeping full memory of learning material, appears a viable approach to learning NLP tasks, and is often superior in generalisation accuracy to eager learning…