Cue Phrase Classi cation Using Machine Learning

Cue phrases may be used in a discourse sense to explicitly signal discourse structure, but also in a sentential sense to convey semantic rather than structural information. Correctly classifying cue phrases as discourse or sentential is critical in natural language processing systems that exploit discourse structure, e.g., for performing tasks such as anaphora resolution and plan recognition. This paper explores the use of machine learning for classifying cue phrases as discourse or sentential. Two machine learning programs (cgrendel and C4.5) are used to induce classi cation models from sets of pre-classi ed cue phrases and their features in text and speech. Machine learning is shown to be an e ective technique for not only automating the generation of classi cation models, but also for improving upon previous results. When compared to manually derived classi cation models already in the literature, the learned models often perform with higher accuracy and contain new linguistic insights into the data. In addition, the ability to automatically construct classication models makes it easier to comparatively analyze the utility of alternative feature representations of the data. Finally, the ease of retraining makes the learning approach more scalable and exible than manual methods.