Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data Export

@misc{Lafferty2001crf, abstract = {We present {\pounds} ¤ ¥ ¦ {\S}¨{\S} ¤ ¥ {\copyright} ��� {\copyright} ¥ ¦ ¤ �� � �� ¦ � , a framework for building probabilistic models to segment and label sequence data. Conditional random fields offer several advantages over hidden Markov models and stochastic grammars for such tasks, including the ability to relax strong independence assumptions made in those models. Conditional random fields also avoid a fundamental limitation of maximum entropy Markov models (MEMMs) and other discriminative Markov models based on directed graphical models, which can be biased towards states with few successor states. We present iterative parameter estimation algorithms for conditional random fields and compare the performance of the resulting models to HMMs and MEMMs on synthetic and natural-language data. 1.}, author = {Lafferty, John and Mccallum, Andrew and Pereira, Fernando}, citeulike-article-id = {3338662}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.9824}, keywords = {conditional, field, random}, posted-at = {2008-09-26 06:25:52}, priority = {2}, title = {Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.9824}, year = {2001} }

TY - ELEC ID - Lafferty2001crf L3 - citeulike-article-id:3338662 N2 - We present £ ¤ ¥ ¦ §¨§ ¤ ¥ © ��� © ¥ ¦ ¤ �� � �� ¦ � , a framework for building probabilistic models to segment and label sequence data. Conditional random fields offer several advantages over hidden Markov models and stochastic grammars for such tasks, including the ability to relax strong independence assumptions made in those models. Conditional random fields also avoid a fundamental limitation of maximum entropy Markov models (MEMMs) and other discriminative Markov models based on directed graphical models, which can be biased towards states with few successor states. We present iterative parameter estimation algorithms for conditional random fields and compare the performance of the resulting models to HMMs and MEMMs on synthetic and natural-language data. 1. TI - Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data KW - conditional KW - field KW - random AU - Lafferty, John AU - Mccallum, Andrew AU - Pereira, Fernando PY - 2001/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.18.9824 ER -