An Introduction to Conditional Random Fields for Relational Learning

By Lise Getoor, B. Taskar. 2007

335 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/24d730b0d8fea050159f272748e39363443a39ce

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

BANNER: An Executable Survey of Advances in Biomedical Named Entity Recognition (2007, background): “In a CRF, each input x i from the sequence of input tokens x = (x 1 … x n ) is a vector of real-valued features or descriptive characteristics, for example, the part of speech.”
Learning Conditional Random Fields for Stereo (2007, methodology): “In the linear models commonly used in language processing, the feature expectations required for gradient-based CRF optimization can be computed efficiently [22].”
Joint Entity Recognition and Disambiguation (2015, methodology): “(Sutton and McCallum, 2006) The parameter estimation algorithm is abstracted in Algorithm 1.”
An Analysis of Active Learning Strategies for Sequence Labeling Tasks (2008, methodology): “We implement all fifteen query selection strategies described in Section 3 for use with CRFs, and evaluate them on all eight data sets.”
Equivalence of Generative and Log-Linear Models (2011, background): “Linear-chain conditional random fields (CRFs)[4], [5] are the log-linear counterpart of Markov chains, e.g., the part-of-speech bigram tagging model.”
Elementary: Large-Scale Knowledge-Base Construction via Machine Learning and Statistical Inference (2012, background): “…mentions in text of the entity Barack Obama. machine learning models such as logistic regression and conditional random fields (Lafferty et al., 2001; Sutton & McCallum, 2006) to learn model parameters from training examples, e.g., annotations of sentences that mention a HasSpouse relationship.”
A conditional random field-based model for joint sequence segmentation and classification (2013, methodology): “In the second level, classification of the observed sequences is performed.”
Structured prediction models for RNN based sequence labeling in clinical text (2016, methodology): “Due to these skip edges, the skip chain CRF model (Sutton and McCallum, 2006) explicitly models dependencies between labels which might be more than one positions apart.”
Structured Models for Fine-to-Coarse Sentiment Analysis (2007, methodology): “Furthermore, these systems have tackled the problem at different levels of granularity, from the document level (Pang et al., 2002), sentence level (Pang and Lee, 2004; Mao and Lebanon, 2006), phrase level (Turney, 2002; Choi et al., 2005), as well as the speaker level in debates (Thomas et al., 200”
Structure-Aware Review Mining and Summarization (2010, background): “The conditional probability is defined as: where Z(x) is the normalization factor, (cid:23) (cid:6) is the state function on node, (cid:18) (cid:6) is the transition functions on edge, and ¸ (cid:28) (cid:6) and (cid:22) (cid:6) are parameters to estimate (Sutton and McCallum, 2006).”