Resumen de TBL template selection: an evolutionary approach

Ruy Luiz Milidiú, Julio Cesar Duarte, Cícero Nogueira dos Santos

• Transformation Based Learning (TBL) is an intensively Machine Learning algorithm frequently used in Natural Language Processing. 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 selection process. We show some empirical evidence that our approach provides template sets with almost the same quality as human built templates.