In particular, the issue of what is a linguistically motivated way of deriving a parser from principle-based theories of grammar is explored. This paper investigates a generative history-based parsing model that synchronises the derivation of non-planar graphs representing semantic depen- dencies with the derivation of dependency trees representing syntactic structures. Chapter 16: ''Verb sense and verb subcategorization probabilities'' by Douglas Roland and Daniel Jurafsky. We look at the well-known case of the ambiguity be- tween a main clause and reduced relative con- struction. We compare different syntactic proposals on the position of numerals in the noun phrase.
James Huang, and James Myers Eds. Brenier, Dan Jurafsky, Wayne Ward, and William Byrne. This corpus-driven computational study addresses the question of why some verbs in some languages participate in the causative alternation while their counterparts in other languages do not. Spitkovsky, Daniel Jurafsky, and Hiyan Alshawi. We demonstrate the benefits of a multilingual approach to automatic lexical semantic verb classification based on statistical analysis of corpora in multiple languages. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in.
Baby Steps: How Less is More in Unsupervised Dependency Parsing. This paper proposes that the set of frequencies that the human language processor keeps track of are those that are useful to it in learning. This book is the fourth in a line of such collections, and its breadth of coverage should make it suitable both as an overview of the state of the field for graduate students, and as a reference for established researchers in Computational Linguistics, Artificial Intelligence, Computer Science, Language Engineering, Information Science, and Cognitive Science. We thus demonstrate an effective combination of deeper linguistic knowledge with the robustness and scalability of statistical techniques. Methods that re-use existing mono-lingual semantic annotation resources to annotate a new language rely on the hypothesis that the semantic annotation scheme used is cross-lingually valid.
We reach an accuracy of 86. We identify a number of non-matching predicates and examine them in a qualitative analysis. Rion Snow, Brendan O'Connor, Daniel Jurafsky and Andrew Y. Yun-Hsuan Sung, Constantinos Boulis, and Dan Jurafsky. In the previous chapter, I have illustrated how configurations and categorial information are encoded in the parser. Narayanan, Srini, and Daniel Jurafsky.
Therefore, an in- vestigation of the quantitative correlates of abstract linguistic concepts might in fact enhance the goals of generative grammar by elucidating the theo- retical relationship between structure and frequency. W eighted influences on the activation competition are possible with other verbs, and the model can therefore also explain data that dem onstrate contextual effects on reduced relatives with sim ple transitives. Knowlege-Free Induction of Morphology using Latent Semantic Analysis. We find that a merge position of numerals higher than adjectives has better results in both methods. What distinguishes our pro-posal from other attempts in function or seman-tic role labelling is that we perform the learning of function labels at the same time as parsing. Frank, Dan Jurafsky and Christopher Potts. Chapter 14: ''Predicting thematic role assignments in context'' by Gerry T.
Rion Snow, Sushant Prakash, Daniel Jurafsky, and Andrew Y. From Baby Steps to Leapfrog: How Less is More in Unsupervised Dependency Parsing. Modular architectures and statistical mechanisms: The case from lexical category disambiguation by Crocker, Matthew ; 11. Pradhan, Sameer, Kadri Hacioglu, Wayne Ward, James H. In this paper, we compare two annotation schemes, Prop- Bank and VerbNet, in a task-independent, general way, analysing how well they fare in capturing the linguistic generalisations that are known to hold for semantic role labels, and consequently how well they grammaticalise aspects of meaning. See the for more details.
We demonstrate that Chinese data is indeed useful in helping to classify the English verbs at 82% accuracy , and furthermore that a multilingual combination of data outperforms the English data alone 85% accuracy. Second, I perform some correlations with sentence completion and sentence production studies, which show that the two data collection methods are not strongly correlated with the corpus counts. Cross-lingual parallelism and small-scale language variation have recently become subject of research in both computational and theoretical linguistics. Michael Levin, Stefan Krawczyk, Steven Bethard, and Dan Jurafsky. Hamilton, Jure Leskovec, Dan Jurafsky. Jurafsky, Daniel, Rebecca Bates, Noah Coccaro, Rachel Martin, Marie Meteer, Klaus Ries, Elizabeth Shriberg, Andreas Stolcke, Paul Taylor, and Carol Van Ess-Dykema. Journal of Memory and Language 60:1, 92-111.
Specifically, we experiment with three lexical semantic classes of English verbs. Viterbi Training Improves Unsupervised Dependency Parsing. A parsing model is defined, that uses the hierarchical mixed-grain representation compiled in a tabular form , thus benefitting from both general and specific information in different stages of processing. Gildea, Daniel and Daniel Jurafsky. Especially important is knowledge about verbs, which are the primary source of relational information in a sentence - the predicate-argument structure that relates an action or state to its participants i. In the experimental study of how much verb guidance occurs in parsing syntactic ambiguity, the preference of an ambiguously subcategorized verb for its continuations is a relevant factor in the choice of the stimuli and in the interpretation of the results. A detailed analysis of the performance of the algorithm and of its errors confirms that the proposed features capture properties related to the argument structure of the verbs.
This paper shows that corpus-based statistical patterns do not entirely concur with human interpretation preferences. We demonstrate the benefits of a multi- lingual approach to automatic lexical se- mantic verb classification based on sta- tistical analysis of corpora in multiple languages. Yun-Hsuan Sung and Dan Jurafsky. The Language of Food: A Linguist Reads the Menu. Feature assignment can be strictly local to one of the configurations defined by X̄ theory, or it might involve sharing features between distant elements.
Measuring Machine Translation Quality as Semantic Equivalence: A Metric Based on Entailment Features. Chapter 3: ''Optimality-theoretic Lexical Functional Grammar'' by Mark Johnson. Knowlege-Free Induction of Inflectional Morphologies. Distributions of four grammatical features are sufficient to reduce error rate by 50% over chance. Rob Voigt and Dan Jurafsky. In this paper, we propose a novel solution to the problem of assigning function labels to syntac-tic constituents. Daniel Cer, Marie-Catherine de Marneffe, Daniel Jurafsky and Christopher D.