Resumen de Adapting Text Simplification Decisions to Different Text Genres and Target Users

Sanja Stajner, Horacio Saggion

• español

  Hemos analizado las alineaciones a nivel de oracion de dos corpus paralelos de textos originales y sus simpli caciones creados con diferentes objetivos. Hemos clasi cado las alineaciones que se observan y dise~nado un algoritmo de clasi cacion capaz de predecir si las oraciones de un texto seran eliminadas, segmentadas, o transformadas durante el proceso de simpli cacion. Hemos realizado una evaluacion cruzada en cada uno de los corpus as como una evaluacion en la cual se entrena en algoritmo en un corpus y se lo evalua en el otro

• English

  We investigate sentence deletion and split decisions in Spanish text simpli cation for two di erent corpora aimed at di erent groups of users. We analyse sentence transformations in two parallel corpora of original and manually simpli ed texts for two di erent types of users and then conduct two classi cation experiments:

  classifying between those sentences to be deleted and those to be kept; and classifying between sentences to be split and those to be left unsplit. Both experiments were rst run on each of the two corpora separately and then run by using one corpus for the training and the other for testing. The results indicated that both sentence decision systems could be successfully trained on one corpus and then used for a di erent text genre in a text simpli cation system aimed at a di erent target population