Active Hidden Markov Models for Information Extraction

By Tobias Scheffer, Christian Decomain, Stefan Wrobel. 2001

675 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

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

In Information Systems

Citations in the corpus

Replay in Deep Learning: Current Approaches and Missing Biological Elements (2021, neutral): “…Wang, Zhang, Li, Zhang, & Lin, 2016; Settles, 2009; Yoo & So Kweon, 2019; Wei, Iyer, & Bilmes, 2015), with the most common selection methods using uncertainty sampling (Lewis & Gale, 1994; Culotta & McCallum, 2005; Scheffer, Decomain, & Wrobel, 2001; Dagan & Engelson, 1995; Dasgupta & Hsu, 2008).”
An Analysis of Active Learning Strategies for Sequence Labeling Tasks (2008, methodology): “A few methods have been proposed, based mostly on the conventions of uncertainty sampling, where the learner queries the instance about which it has the least certainty (Scheffer et al., 2001; Culotta and McCallum, 2005; Kim et al., 2006), or query-by-committee, where a “committee” of models selects”
A Survey of Deep Active Learning (2020, background): “For example, in the margin sampling [188], margin M is defined as the difference between the predicted highest probability and the predicted second highest probability of an sample as follows: M = P (y1 | x ) − P (y2 | x ) , where y1 and y2 are the first and second most probable labels predicted for”
Active Learning for Efficient Audio Annotation and Classification with a Large Amount of Unlabeled Data (2019, methodology): “In the binary classification case, it is equivalent to two other common certainty-based sampling: maximum entropy [24] and smallest margin [25].”
Reducing Labeling Effort for Structured Prediction Tasks (2005, background): “…estimates in natural language tasks have been studied in domains such as text classiﬁcation (Gandrabur & Foster 2003), information extraction (Scheffer, Decomain, & Wrobel 2001; Culotta & McCallum 2004), and speech recognition (Gunawardana, Hon, & Jiang 1998), although none of these consider…”
SeqMix: Augmenting Active Sequence Labeling via Sequence Mixup (2020, background): “So far, various uncertainty-based (Scheffer et al., 2001; Culotta and McCallum, 2005; Kim et al., 2006) and committee-based approaches (Dagan and Engelson, 1995) 2017) further improve the aforementioned active learning approaches to improve the sampling diversity as well as the model’s…”
Interactive Information Extraction with Constrained Conditional Random Fields (2004, background): “For example, Scheffer, Decomain, & Wrobel (2001) derive confidence estimates using hidden Markov models in an information extraction system, however, they do not estimate the confidence of entire fields, only singleton tokens.”
A Holistic Approach to Undesired Content Detection in the Real World (2022, neutral): “…sampling selects data about which the model is most uncertain, which can be quantified by predicted probabilities (Lewis and Catlett 1994; Scheffer, Decomain, and Wrobel 2001), disagreement among an ensemble of models (Seung, Opper, and Sompolinsky 1992; McCallum and Nigam 1998), or by…”
Cost-Effective Active Learning for Deep Image Classification (2017, methodology): “, LC [31], MS [32], and EN [33] to select K most informative/uncertain samples left in DU .”
Active Learning Literature Survey (2009, background): “Most of these algorithms can be generalized to other probabilistic sequence models, such as HMMs (Dagan and Engelson, 1995; Scheffer et al., 2001) and probabilistic context-free grammars (Baldridge and Osborne, 2004; Hwa, 2004).”