AbstractThis paper separates conditional parameter estima- tion, which consistently raises test set accuracy on statistical NLP tasks, from conditional model struc- tures, such as the conditional Markov model used for maximum-entropy tagging, which tend to lower accuracy. Error analysis on part-of-speech tagging shows that the actual tagging errors made by the conditionally structured model derive not only from label bias, but also from other ways in which the independence assumptions of the conditional model structure are unsuited to linguistic sequences. The paper presents new word-sense disambiguation and POS tagging experiments, and integrates apparently conﬂicting reports from other recent work.
Similar books and articles
A Mental Model Analysis of Young Children's Conditional Reasoning with Meaningful Premises.Henry Markovits - 2000 - Thinking and Reasoning 6 (4):335 – 347.
Conditionals, Comparative Probability, and Triviality: The Conditional of Conditional Probability Cannot Be Represented in the Object Language.Charles G. Morgan - 1999 - Topoi 18 (2):97-116.
Conditionals and Conditional Thinking.Andrea Manfrinati, Pierdaniele Giaretta & Paolo Cherubini - 2008 - Mind and Society 7 (1):21-34.
Conjunction, Disjunction and Iterated Conditioning of Conditional Events.Angelo Gilio & Giuseppe Sanfilippo - 2013 - In R. Kruse (ed.), Advances in Intelligent Systems and Computing. Springer.
Remarks on the Theory of Conditional Probability: Some Issues of Finite Versus Countable Additivity.Teddy Seidenfeld - unknown
The Causal Structure of Utility Conditionals.Jean-François Bonnefon & Steven A. Sloman - 2013 - Cognitive Science 37 (1):193-209.
Iterated Revision and Minimal Change of Conditional Beliefs.Craig Boutilier - 1996 - Journal of Philosophical Logic 25 (3):263 - 305.
Added to PP
Historical graph of downloads
Citations of this work
Feature-Rich Part-of-Speech Tagging with a Cyclic Dependency Network.Christopher Manning - manuscript
References found in this work
No references found.