Structure Learning for Language Processing
Yizhao Ni, Craig Saunders, Sandor Szedmak, Mahesan Niranjan
We applied a structure learning model, max-margin structure, to natural language processing tasks, where the aim is to capture the latent relationships within the output language domain. We formulate this model as an extension of multi?class SVM and present a perceptron?based learning approach to solve the problem. Experiments are carried out on two related NLP tasks: part?of?speech tagging and machine translation, illustrating the effectiveness of the model.
MLSP 2009 (IEEE Workshop on Machine Learning for Signal Processing), Grenoble France, 2-4 Septembre, 2009