An adaptive statistical languagemodel is described, which successfullyintegrates long distancelinguistic information with other knowledge sources. Most existing statistical language models exploit only the immediate history of a text. To extract information from further back in the document's history, we propose and use trigger pairs as the basic information bearing elements. This allows the model to adapt its expectations to the topic of discourse. Next, statistical evidence from multiple sources must be combined. Traditionally, linear interpolation and its variants have been used, but these are shown here to be seriously deficient. Instead, we apply the principle of Maximum Entropy (ME). Each information source gives rise to a set of constraints, to be imposed on the combined estimate. The intersection of these constraints is the set of probability functions which are consistent with all the information sources. The function with the highest entropy within that set is the ME solution...

Most probabilistic classifiers used for word-sense disambiguation have either been based on only one contextual feature or have used a model that is simply assumed to characterize the interdependencies among multiple contextual features. In this paper, a different approach to formulating a probabilistic model is presented along with a case study of the performance of models produced in this manner for the disambiguafion of the noun interest. We describe a method for formulating probabilistic models that use multiple contextual features for word-sense disambiguafion, without requiring untested assumptions regarding the form of the model. Using this approach, the joint distribution of all variables is described by only the most systematic variable interactions, thereby limiting the number of parameters to be estimated, supporting computational efficiency, and providing an understanding of the data.

In this paper, we investigate a new statistical language model which captures topic-related dependenciesof words within and across sentences. First, we develop a sentence-level mixture language model that takes advantage of the topic constraints in a sentence or article. Second, we introduce topic-dependent dynamic cache adaptation techniques in the framework of the mixture model. Experiments with the static (or unadapted) mixture model on the 1994 WSJ task indicated a 21% reduction in perplexity and a 3-4% improvement in recognition accuracy over a general n-gram model. The static mixture model also improved recognition performance over an adapted n-gram model. Mixture adaptation techniques contributed a further 14% reduction in perplexity and a small improvement in recognition accuracy.

Abstract—In certain contexts, maximum entropy (ME) modeling can be viewed as maximum likelihood (ML) training for exponential models, and like other ML methods is prone to overfitting of training data. Several smoothing methods for ME models have been proposed to address this problem, but previous results do not make it clear how these smoothing methods compare with smoothing methods for other types of related models. In this work, we survey previous work in ME smoothing and compare the performance of several of these algorithms with conventional techniques for smoothing-gram language models. Because of the mature body of research in-gram model smoothing and the close connection between ME and conventional-gram models, this domain is well-suited to gauge the performance of ME smoothing methods. Over a large number of data sets, we find that fuzzy ME smoothing performs as well as or better than all other algorithms under consideration. We contrast this method with previous-gram smoothing methods to explain its superior performance. Index Terms—Exponential models, language modeling, maximum entropy, minimum divergence,-gram models, smoothing.

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...

Language models play an important role in improving the accuracy of a continuous speech recognizer. In this thesis, we introduce a new statistical language model which captures long term topic dependencies of words within and across sentences. The model includes two main contributions. First, we develop a topic-dependent sentence-level mixture language model which takes advantage of the topic constraints in a sentence or a paragraph. Since this language model is not Markov and has a large search space, it is used only in the last stage of a multi-pass search strategy in the recognizer. Second, we introduce topic-dependent dynamic adaptation techniques in the framework of the mixture model. During the course of this thesis, we also investigate robust parameter estimation techniques, which are extremely important in light of the sparse data problems in language modeling. The model is implemented in the BU speech recognition system and provides a significant improvement in recognition accuracy. An important advantage of the framework of our model is that it is a simple extension of existing language modeling techniques that can easily be integrated with other language modeling advances.

Many of the kinds of language model used in speech understanding suffer from imperfect modeling of intra-sentential contextual influences. I argue that this problem can be addressed by clustering the sentences in a training corpus automatically into subcorpora on the criterion of entropy reduction, and calculating separate language model parameters for each cluster. This kind of clustering offers a way to represent impor- tant contextual effects and can therefore significantly improve the performance of a model. It also offers a reasonably automatic means to gather evidence on whether a more complex, context-sensitive model using the same general kind of linguistic information is likely to reward the effort that would be required to develop it: if clustering improves the performance of a model, this proves the existence of further context dependencies, not exploited by the unclustered model. As evidence for these claims, I present results showing that clustering improves some models but not others for the ATIS domain. These results are consistent with other findings for such models, suggesting that the existence or otherwise of an improvement brought about by clustering is indeed a good pointer to whether it is worth developing further the unclustered model.

There are two main topics in this thesis, a corpus-based parser and a study of sublanguage. A novel approach to corpus-based parsing is proposed. In this framework, a probabilistic grammar is constructed whose rules are partial trees from a syntacticallybracketed corpus. The distinctive feature is that the partial trees are multi-layered. In other words, only a small number of non-terminals are used to cut the initial trees; other grammatical nodes are embedded into the partial trees, and hence into the grammar rules. Good parsing performance was obtained, even with small training corpora. Several techniques were developed to improve the parser's accuracy, including in particular two methods for incorporating lexical information. One method uses probabilities of binary lexical dependencies; the other directly lexicalizes the grammar rules. Because the grammar rules are long, the number of rules is huge -- more than thirty thousand from a corpus of one million words. A parsing algorithm...

this document. The first section, in chapter 3, develops a model for syntactic dependencies based on word-category n-grams. The second section, in chapter 4, extends this model by allowing short-range word relations to be captured through the incorporation of selected word n-grams.

Standard statistical language models, or n-gram models, which represent the probability of word sequences, suffer from sparse-data problems in tasks where large amounts of domain-specific text are not available. This thesis focuses on improving the estimation of domain-dependent n-gram models by using out-of-domain text data. Previous approaches for estimating language models from multi-domain data have not accounted for the characteristic variations of style and content across domains. In contrast, this thesis introduces two approaches that compensate for multi-domain differences, both representing "style" by part-of-speech (POS) sequences and "content" by the particular choice of words. First, data from multiple domains is combined using similarity weighting schemes that discriminate for content and style relevance prior to pooling multi-domain text. Second, n-gram distributions from multiple domains are combined, via a POS-dependent n-gram framework that separately compensate for word and POS usage differences. Two variations are explored: explicitly transforming the out-of-domain distribution before combining with an in-domain model, and vi separately estimating components of the POS-dependent n-gram model using multidomain data. Finally, measures to analyze and predict recognition performance of language models are also investigated, resulting in an algorithm for predicting performance differences associated with localized changes in language models given a recognition system.