Semi-Markov Conditional Random Fields for Information Extraction

By Sunita Sarawagi, William W. Cohen. 2004

756 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Reasoning about Quantities in Natural Language (2015, methodology): “We compare the accuracy and running times of the Semi-CRF model (SC) (Sarawagi and Cohen, 2004) and the bank of classifiers model (C+I) (PR) (Punyakanok and Roth, 2001), using 10-fold cross-validation.”
Latent Predictor Networks for Code Generation (2016, methodology): “α and β can be computed efficiently using the forward-backward algorithm for Semi-Markov models [12], where we associate P (rt | y1.”
Search-based structured prediction (2009, neutral): “In our experiments, we focus exclusively on chunk-at-a-time decoding, as it is more expressive (feature-wise) and has been seen to perform better in other scenarios (Sarawagi and Cohen 2004).”
Anticipating Human Activities Using Object Affordances for Reactive Robotic Response (2013, methodology): “Such frameworks as HMMs [53], [54], DBNs [55], CRFs [5], [56], [57], semi-CRFs [58] have been previously used to model the temporal structure of videos and text.”
Protein Fold Recognition Using Segmentation Conditional Random Fields (SCRFs) (2006, neutral): “…involved within the fold, i.e., E 2 = ∅ , SCRFs degrade into semi-Markov CRF models, which are linear CRFs that allow observations of variable length (Sarawagi and Cohen, 2004); on the other hand, if the state transitions between neighboring segments are deterministic, we do not need to consider…”
Joint Entity Recognition and Disambiguation (2015, methodology): “Inspired by the forward backward algorithm (Sha and Pereira, 2003) and Semi-CRF (Sarawagi and Cohen, 2004), we leverage dynamic programming techniques to compute the normalization factor Z w and marginal probability P ( a j | x t , w ) when w is given.”
An Introduction to Conditional Random Fields (2010, background): “Semi-Markov CRFs [105] add somewhat more flexibility in choosing features, by allowing features functions to depend on larger segments of the input that depend on the output labelling.”
Identifying Sources of Opinions with Conditional Random Fields and Extraction Patterns (2005, methodology): “CRFs have been used successfully for Named Entity recognition (e.g., McCallum and Li (2003), Sarawagi and Cohen (2004)), and AutoSlog has performed well on information extraction tasks in several domains (Riloff, 1996a).”
Mining knowledge from text using information extraction (2005, methodology): “Several earlier IE systems used generative HMM models [4; 31]; however, discriminately-trained CRF models have recently been shown to have an advantage over HMM’s [54; 65].”
A Unified Generative Framework for Aspect-based Sentiment Analysis (2021, methodology): “Xu et al. (2020) design the position-aware tagging schema and apply model based on CRF (Lafferty et al., 2001) and Semi-Markov CRF (Sarawagi and Cohen, 2004).”
More Data, More Relations, More Context and More Openness: A Review and Outlook for Relation Extraction (2020, methodology): “Graphical methods (Roth and Yih, 2002, 2004; Sarawagi and Cohen, 2005; Yu and Lam, 2010) abstract the dependencies between entities, text and relations in the form of directed acyclic graphs, and then use infer-”
Semi-Supervised Learning for Natural Language (2005, methodology): “By using semi-Markov Conditional Random Fields, (Cohen and Sarawagi, 2004; Sarawagi and Cohen, 2004) recast the named-entity recognition problem as a segmentation problem rather than a tagging problem.”
Integer linear programming inference for conditional random fields (2005, methodology): “01) and the discriminative method (based on voted perceptron, suggested by Sha and Pereira (2003); Collins (2002)).”
Parsing Videos of Actions with Segmental Grammars (2014, methodology): “Previous work has used semi-Markov models to either capture temporal dependencies between actions [23, 27, 9] or compositional dependencies between subactions within an action [17, 26].”
Global inference for sentence compression : an integer linear programming approach (2008, methodology): “The algorithm is an extension of Viterbi for the case in which scores factor over dynamic sub-strings ( Sarawagi & Cohen, 2004; McDonald, Crammer, & Pereira, 2005a).”
SCARF: A Segmental CRF Speech Recognition System (2009, methodology): “The SCARF system uses Segmental Conditional Random Fields - also known as Semi-Markov Random Fields [3] or SCRFs - as a theoretical underpinning.”
An Introduction to Conditional Random Fields for Relational Learning (2007, background): “Semi-Markov CRFs [Sarawagi and Cohen, 2005] add somewhat more ﬂexibility in choosing features, which may be useful for certain tasks in information extraction and especially bioinformatics.”
A segmental CRF approach to large vocabulary continuous speech recognition (2009, neutral): “SEGMENTAL CRF FORMULATION Segmental Conditional Random Fields -also known as Semi-Markov Random Fields [16] or SCRFs -underlie our approach.”
Segmental Spatiotemporal CNNs for Fine-Grained Action Segmentation (2016, methodology): “1: The forward pass of our segmental inference algorithm compared with Segmental Viterbi [18].”