Open Access

Evolutionary TBL template generation

  • Ruy Luiz Milidiú1Email author,
  • Julio Cesar Duarte1, 2 and
  • Cícero Nogueira dos Santos1
Journal of the Brazilian Computer Society13:BF03194255

https://doi.org/10.1007/BF03194255

Abstract

Transformation Based Learning (TBL) is a Machine Learning technique frequently used in some Natural Language Processing (NLP) tasks. TBL uses rule templates to identify error-correcting patterns. A critical requirement in TBL is the availability of a problem domain expert to build these rule templates. In this work, we propose an evolutionary approach based on Genetic Algorithms to automatically implement the template generation process. Additionally, we report our findings on five experiments with useful NLP tasks. We observe that our approach provides template sets with a mean loss of performance of 0.5% when compared to human built templates

Keywords

Machine Learning Genetic Algorithms Transformation Error-Driven Based Learning