Part-of-Speech Tagging with Two Sequential Transducers
André Kempe · 2001
We describe a memory-based classification architecture for word sense disambiguation and our experience with its application to the SENSEVAL evaluation task. In a memory-based approach, selecting the correct sense of a word in a new context is achieved by finding the closest match to stored examples of this task. Advantages of the approach include (i) fast development time for classifiers, (ii) easy and elegant automatic integration of information sources, (iii) use of all available data, and (iv) relatively high accuracy without language engineering.