Study on interaction between entropy pruning and kneser-ney smoothing

By Ciprian Chelba, Thorsten Brants, Will Neveitt, Peng Xu. 2010

44 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing, Signal Processing

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

1. Brian Hutchinson
2. Mari Ostendorf
3. Maryam Fazel
4. Li Deng
5. Dong Yu

In Natural Language Processing

1. Brian Hutchinson
2. Mari Ostendorf
3. Maryam Fazel

In Signal Processing

1. Brian Hutchinson
2. Mari Ostendorf
3. Maryam Fazel
4. Li Deng
5. E.J. Candes
6. B. Recht

Citations in the corpus

• One billion word benchmark for measuring progress in statistical language modeling (2013, neutral): “Since in practice these models are pruned, often quite aggressivley, we also illustrate the negative effect of (Stolcke, 1998) entropy pruning on both models, similar to (Chelba et al., 2010).”
• Low Rank Language Models for Small Training Sets (2011, methodology): “Empirical studies show that the modified Kneser–Ney method works well over a range of training set sizes on a variety of sources [1], although other methods are more effective when pruning is used in training large language models [2].”
• The A2iA Arabic Handwritten Text Recognition System at the Open HaRT2013 Evaluation (2014, unknown): “However, in [18], it has been shown that such pruning techniques with a KneserNey smoothed LM might results in a poor language model quality.”