We demonstrate that an unlexicalized PCFG can parse much more accurately than previously shown, by making use of simple, linguistically motivated state splits, which break down false independence assumptions latent in a vanilla treebank grammar. Indeed, its performance of 86.36% (LP/LR F1) is better than that of early lexicalized PCFG models, and surprisingly close to the current state-of-the-art. This result has potential uses beyond establishing a strong lower bound on the maximum possible accuracy of unlexicalized models: an unlexicalized PCFG is much more compact, easier to replicate, and easier to interpret than more complex lexical models, and the parsing algorithms are simpler, more widely understood, of lower asymptotic complexity, and easier to optimize.

/pdf/accurate-unlexicalized-parsing-5azxkuoluw.pdf

Accurate Unlexicalized Parsing

This article describes three statistical models for natural language parsing. The models extend methods from probabilistic context-free grammars to lexicalized grammars, leading to approaches in which a parse tree is represented as the sequence of decisions corresponding to a head-centered, top-down derivation of the tree. Independence assumptions then lead to parameters that encode the X-bar schema, subcategorization, ordering of complements, placement of adjuncts, bigram lexical dependencies, wh-movement, and preferences for close attachment. All of these preferences are expressed by probabilities conditioned on lexical heads. The models are evaluated on the Penn Wall Street Journal Treebank, showing that their accuracy is competitive with other models in the literature. To gain a better understanding of the models, we also give results on different constituent types, as well as a breakdown of precision/recall results in recovering various types of dependencies. We analyze various characteristics of the models through experiments on parsing accuracy, by collecting frequencies of various structures in the treebank, and through linguistically motivated examples. Finally, we compare the models to others that have been applied to parsing the treebank, aiming to give some explanation of the difference in performance of the various models.

/pdf/head-driven-statistical-models-for-natural-language-parsing-l1eadq1q9u.pdf

Head-Driven Statistical Models for Natural Language Parsing

In human sentence processing, cognitive load can be defined many ways This report considers a definition of cognitive load in terms of the total probability of structural options that have been disconfirmed at some point in a sentence: the surprisal of word wi given its prefix woi-1 on a phrase-structural language model These loads can be efficiently calculated using a probabilistic Earley parser (Stolcke, 1995) which is interpreted as generating predictions about reading time on a word-by-word basis Under grammatical assumptions supported by corpus-frequency data, the operation of Stolcke's probabilistic Earley parser correctly predicts processing phenomena associated with garden path structural ambiguity and with the subject/object relative asymmetry

A probabilistic earley parser as a psycholinguistic model

Presents the theory, design principles, implementation and performance results of PicHunter, a prototype content-based image retrieval (CBIR) system. In addition, this document presents the rationale, design and results of psychophysical experiments that were conducted to address some key issues that arose during PicHunter's development. The PicHunter project makes four primary contributions to research on CBIR. First, PicHunter represents a simple instance of a general Bayesian framework which we describe for using relevance feedback to direct a search. With an explicit model of what users would do, given the target image they want, PicHunter uses Bayes's rule to predict the target they want, given their actions. This is done via a probability distribution over possible image targets, rather than by refining a query. Second, an entropy-minimizing display algorithm is described that attempts to maximize the information obtained from a user at each iteration of the search. Third, PicHunter makes use of hidden annotation rather than a possibly inaccurate/inconsistent annotation structure that the user must learn and make queries in. Finally, PicHunter introduces two experimental paradigms to quantitatively evaluate the performance of the system, and psychophysical experiments are presented that support the theoretical claims.

/pdf/the-bayesian-image-retrieval-system-pichunter-theory-4rqjnp3alm.pdf

The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments

In this paper we first propose a new statistical parsing model, which is a generative model of lexicalised context-free grammar. We then extend the model to include a probabilistic treatment of both subcategorisation and wh-movement. Results on Wall Street Journal text show that the parser performs at 88.1/87.5% constituent precision/recall, an average improvement of 2.3% over (Collins 96).

/pdf/three-generative-lexicalised-models-for-statistical-parsing-5aglnw17py.pdf

Three Generative, Lexicalised Models for Statistical Parsing

The problem of assigning a probability to each word of a language is considered. Two methods are discussed. One method assigns a probability to a word on the basis of particular measurable features of the language. The second method is applied to languages L(G) generated by a grammar G. A probability is associated with each production of G. These in turn define the word probabilities of each word in the language. The conditions for this assignment to be a probabilistic measure are derived.

Applying Probability Measures to Abstract Languages

The nature of probability-measure languages (pm-languages) has been investigated [2], [3], [7], [8], in particular, those languages generated by given probabilistic grammars (p-grammars). However, the determination of a p-grammar that can generate some given language has been an open question. Since languages are infinite in general, the specification of a pm-language is vague. In this paper, it is assumed that some finite representation exists for the set of words (this can be a nonprobabilistic grammar) and that the probability of each word in the language is computable by some word function whose domain is the language.

R.A. Thompson

Papers

Applying Probability Measures to Abstract Languages

Determination of Probabilistic Grammars for Functionally Specified Probability-Measure Languages