Supervised Sequence Segmentation Task
- See: Sequence Segmentation Statistical Models.
- (Sarawagi, 2006) ⇒ Sunita Sarawagi. (2006). “Efficient Inference on Sequence Segmentation Models.” In: Proceedings of the 23rd International Conference on Machine Learning (ICML 2006). doi:10.1145/1143844.1143944
- Given an input sequence x = x1, ..., xn, a segmentation s of x consists of a sequence of variable length segments s = s1, ..., sp where each segment sj = <tj, uj, yj> consists of a start position tj, an end position uj, and a label yj ∈ [math]Y[/math] . Conceptually, a segment means that the tag yj is given to all xi’s between [math]i[/math] = tj and [math]i[/math] = uj, inclusive. Each segment sj can be associated with a vector of features that captures the dependence of its label on input properties in the neighborhood of the segment and the label of the segment before it. The goal during inference is to simultaneously find a segmentation of the input sequence and label each segment so as to maximize the total score over all segments.
- (Keshet et al., 2005) ⇒ J. Keshet, B. Shalev-Shwartz, and Yoram Singer. (2005). “Phoneme alignment using large margin techniques.” In: Proceedings of the NIPS 2005 Workshop on the Advances in Structured Learning for Text and Speech Processing.
- (McDonald et al., 2005) ⇒ Ryan T. McDonald, Koby Crammer, and Fernando Pereira. (2005). “Flexible Text Segmentation with Structured Multilabel Classification.” In: Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing (HLT/EMNLP, 2005).
- (Sarawagi & Cohen, 2004) ⇒ Sunita Sarawagi, and William W. Cohen. (2004). “Semi-Markov Conditional Random Fields for Information Extraction.” In: Proceedings of Advances in Neural Information Processing Systems, 17 NIPS 2004.