An Introduction to Conditional Random Fields for Relational Learning

By Charles Sutton, Andrew McCallum. 2007

1,036 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/5969fdc215def657b8e607aaa03cee3796c10d4a

How researchers see this paper

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

Citations in the corpus

Negation scope detection in sentiment analysis: Decision support for news-driven trading (2016, methodology): “For the purpose of negation scope detection, we specify a socalled linear-chain conditional random field with a current observation identity [37].”
Automatic classification of sentences to support Evidence Based Medicine (2011, methodology): “We employ Conditional Random Fields (CRF) [ 5 ], which are well suited for learning over sequential data (such as cohesive, structured text).”
Joint Entity Recognition and Disambiguation (2015, methodology): “(Sutton and McCallum, 2006) The parameter estimation algorithm is abstracted in Algorithm 1.”
Bayesian reasoning and machine learning (2012, unknown): “6 Conditional random fields For an input x and output y, a CRF is defined by a conditional distribution [266, 166] p(y|x) = 1 Z(x) ∏”
A review on vision techniques applied to Human Behaviour Analysis for Ambient-Assisted Living (2012, background): “As a refinement stage, Conditional Random Fields (CRF) [113] are applied to model sequential observations and recognise completed activities among combinations of local on-going activities.”
BANNER: An Executable Survey of Advances in Biomedical Named Entity Recognition (2007, background): “There are several good introductions to conditional random fields, such as [14] and [18].”
Autotagging Facebook: Social network context improves photo annotation (2008, background): “For this inference task, we assume the availability of a face recognition system that accepts a face image and returns a distribution of likelihoods over the label space L ([26] provides a survey of possibilities).”
Automatic prosody prediction and detection with Conditional Random Field (CRF) models (2010, background): “Prosody modeling can be seen as relational learning [ 15 ].”
Text Classification Algorithms: A Survey (2019, background): “Limitation of Conditional Random Field (CRF) With regards to CRF, the most evident disadvantage of CRF is the high computational complexity of the training step [175], especially for text data sets due to high feature space.”
Domain-constrained semi-supervised mining of tracking models in sensor networks (2007, background): “in [10], L(Λ) is concave in light of the convexity of the kind of functions g(x) = log P i exp xi.”
Joint People, Event, and Location Recognition in Personal Photo Collections Using Cross-Domain Context (2010, result): “This is in contrast to standard CRF models [18] that require learning class-speciﬁc feature coeﬃcients for each class.”
Understanding Video Events: A Survey of Methods for Automatic Interpretation of Semantic Occurrences in Video (2009, methodology): “Learning of CRF parameters can be achieved using convex optimization methods such as conjugate gradient descent [138].”
Practical Very Large Scale CRFs (2010, methodology): “In this section, we recall the basics of Conditional Random Fields (CRFs) (Lafferty et al., 2001; Sutton and McCallum, 2006) and introduce the notations that will be used throughout.”