A Cascaded Machine Learning Approach to Interpreting Temporal Expressions

A new architecture for identifying and interpreting temporal expressions is introduced, in which the large set of complex hand-crafted rules standard in systems for this task is replaced by a series of machine learned classifiers and a much smaller set of context-independent semantic composition rules. Experiments with the TERN 2004 data set demonstrate that overall system performance is comparable to the state-of-the-art, and that normalization performance is particularly good.

• Type: Conference
• Year: 2007
• Where: NAACL
• Authors: David Ahn, Joris van Rantwijk, Maarten de Rijke
• Keywords: Complex Hand-crafted Rules, Computational Linguistics, Data Set Demonstrate, NAACL 2007, Semantic Composition Rules