The natural language processing community has recently experienced a growth of interest in domain independent shallow semantic parsing—the process of assigning a WHO did WHAT to WHOM, WHEN, WHERE, WHY,HOW etc. structure to plain text. This process entails identifying groups of words in a sentence that represent these semantic arguments and assigning specific labels to them. It could play a key role in NLP tasks like Information Extraction, Question Answering and Summarization. We propose a machine learning algorithm for semantic role parsing, extending the work of Gildea and Jurafsky (2002), Surdeanu et al. (2003) and others. Our algorithm is based on Support Vector Machines which we show give large improvement in performance over earlier classifiers. We show performance improvements through a number of new features designed to improve generalization to unseen data, such as automatic clustering of verbs. We also report on various analytic studies examining which features are most important, comparing our classifier to other machine learning algorithms in the literature, and testing its generalization to new test set from different genre. On the task of assigning semantic labels to the PropBank (Kingsbury, Palmer, & Marcus, 2002) corpus, our final system has a precision of 84 % and a recall of 75%, which are the best results currently reported for this task. Finally, we explore a completely different architecture which does not requires a deep syntactic parse. We reformulate the task as a combined chunking and classification problem, thus allowing our algorithm to be applied to new languages or genres of text for which statistical syntactic parsers may not be available.

Probabilistic Context-Free Grammars (PCFGs) and variations on them have recently become some of the most common formalisms for parsing. It is common with PCFGs to compute the inside and outside probabilities. When these probabilities are multiplied together and normalized, they produce the probability that any given non-terminal covers any piece of the input sentence. The traditional use of these probabilities is to improve the probabilities of grammar rules. In this thesis we show that these values are useful for solving many other problems in Statistical Natural Language Processing. We give a framework for describing parsers. The framework generalizes the inside and outside values to semirings. It makes it easy to describe parsers that compute a wide variety of interesting quantities, including the inside and outside probabilities, as well as related quantities such as Viterbi probabilities and n-best lists. We also present three novel uses for the inside and outside probabilities. T...

This paper shows how finite approximations of long distance dependency (LDD) resolution can be obtained automatically for wide-coverage, robust, probabilistic Lexical-Functional Grammar (LFG) resources acquired from treebanks. We extract LFG subcategorisation frames and paths linking LDD reentrancies from f-structures generated automatically for the Penn-II treebank trees and use them in an LDD resolution algorithm to parse new text. Unlike (Collins, 1999; Johnson, 2002), in our approach resolution of LDDs is done at f-structure (attribute-value structure representations of basic predicate-argument or dependency structure) without empty productions, traces and coindexation in CFG parse trees. Currently our best automatically induced grammars achieve 80.97 % f-score for fstructures parsing section 23 of the WSJ part of the Penn-II treebank and evaluating against the DCU 105 1 and 80.24 % against the PARC 700 Dependency Bank (King et al., 2003), performing at the same or a slightly better level than state-of-the-art hand-crafted grammars (Kaplan et al., 2004). 1

Interactive spoken dialogue provides many new challenges for natural language understanding systems. One of the most critical challenges is simply determining the speaker’s intended utterances: both segmenting a speaker’s turn into utterances and determining the intended words in each utterance. Even assuming perfect word recognition, the latter problem is complicated by the occurrence of speech repairs, which occur where speakers go back and change (or repeat) something they just said. The words that are replaced or repeated are no longer part of the intended utterance, and so need to be identified. Segmenting turns and resolving repairs are strongly intertwined with a third task: identifying discourse markers. Because of the interactions, and interactions with POS tagging and speech recognition, we need to address these tasks together and early on in the processing stream. This paper presents a statistical language model in which we redefine the speech recognition problem so that it includes the identification of POS tags, discourse markers, speech repairs and intonational phrases. By solving these simultaneously, we obtain better results on each task than addressing them separately. Our model is able to identify 72 % of turn-internal intonational boundaries with a precision of 71%, 97 % of discourse markers with 96 % precision, and detect and correct 66 % of repairs with 74 % precision.

Broad coverage, high quality parsers are available for only a handful of languages. A prerequisite for developing broad coverage parsers for more languages is the annotation of text with the desired linguistic representations (also known as “treebanking”). However, syntactic annotation is a labor intensive and time-consuming process, and it is difficult to find linguistically annotated text in sufficient quantities. In this article, we explore using parallel text to help solving the problem of creating syntactic annotation in more languages. The central idea is to annotate the English side of a parallel corpus, project the analysis to the second language, and then train a stochastic analyzer on the resulting noisy annotations. We discuss our background assumptions, describe an initial study on the “projectability ” of syntactic relations, and then present two experiments in which stochastic parsers are developed with minimal human intervention via projection from English. 1

Building models of language is a central task in natural language processing. Traditionally, language has been modeled with manually-constructed grammars that describe which strings are grammatical and which are not; however, with the recent availability of massive amounts of on-line text, statistically-trained models are an attractive alternative. These models are generally probabilistic, yielding a score reflecting sentence frequency instead of a binary grammaticality judgement. Probabilistic models of language are a fundamental tool in speech recognition for resolving acoustically ambiguous utterances. For example, we prefer the transcription forbear to four bear as the former string is far more frequent in English text. Probabilistic models also have application in optical character recognition, handwriting recognition, spelling correction, part-of-speech tagging, and machine translation. In this thesis, we investigate three problems involving the probabilistic modeling of languag...

Excellent results have been reported for DataOriented Parsing (DOP) of natural language texts (Bod, 1993c). Unfortunately, existing algorithms are both computationally intensive and difficult to implement. Previous algorithms are expensive due to two factors: the exponential number of rules that must be generated and the use of a Monte Carlo p arsing algorithm. In this paper we solve the first problem by a novel reduction of the DOP model toga small, equivalent probabilistic context-free grammar. We solve the second problem by a novel deterministic parsing strategy that maximizes the expected number of correct con- stituents, rather than the probability of a correct parse tree. Using ithe optimizations, experiments yield a 97% crossing brackets rate and 88% zero crossing brackets rate. This differs significantly from the results reported by Bod, and is compara- ble to results from a duplication of Pereira and Schabes's (1992) experiment on the same data. We show that Bod's results are at least partially due to an extremely fortuitous choice of test data, and partially due to using cleaner data than other researchers.

Parser development is generally viewed as a primarily linguistic enterprise. A grammarian examines sentences, skillfully extracts the linguistic generalizations evident in the data, and writes grammar rules which cover the language. The grammarian

This paper presents an algorithm for automatically assigning phrase breaks to unrestricted text for use in a text-to-speech synthesizer. Text is first converted into a sequence of part-of-speech tags. Next a Markov model is used to give the most likely sequence of phrase breaks for the input part-of-speech tags. In the Markov model, states represent types of phrase break and the transitions between states represent the likelihoods of sequences of phrase types occurring. The paper reports a variety of experiments investigating part-of-speech tag-sets, Markov model structure and smoothing. The best setup correctly identifies 79 % of breaks in the test corpus. © 1998 Academic Press Limited 1.

This paper investigates the role of resource allocation as a source of processing difficulty in human sentence comprehension. The paper proposes a simple informationtheoretic characterization of processing difficulty as the work incurred by resource reallocation during parallel, incremental, probabilistic disambiguation in sentence comprehension, and demonstrates its equivalence to the theory of Hale (2001), in which the difficulty of a word is proportional to its surprisal (its negative log-probability) in the context within which it appears. This proposal subsumes and clarifies findings that high-constraint contexts can facilitate lexical processing, and connects these findings to well-known models of parallel constraint-based comprehension. In addition, the theory leads to a number of specific predictions about the role of expectation in syntactic comprehension, including the reversal of locality-based difficulty patterns in syntactically constrained contexts, and conditions under which increased ambiguity facilitates processing. The paper examines a range of established results bearing on these predictions, and shows that they are largely consistent with the surprisal theory.