Articles

The Brown and the Berkeley parsers are two state-of-the-art generative parsers. Since both parsers produce n-best lists, it is possible to apply reranking techniques to the output of both of these parsers, and to their union. We note that the standard reranker feature set dis- tributed with the Brown parser …...

Recent work has proposed the use of an ex- tracted tree grammar as the basis for treebank analysis and search queries, in which queries are stated over the elementary trees, which are small chunks of syntactic structure. However, this work was lacking in two crucial ways. First, it did not …...

The paper describes a data driven dependency parsing approach which uses clausal informa- tion of a sentence to improve the parser per- formance. The clausal information is added automatically during the parsing process. We demonstrate the experiments on Hindi, a lan- guage with relatively rich case marking sys- tem and …...

Stochastic Inversion Transduction Grammars constitute a powerful formalism in Machine Translation for which an efficient Dynamic Programming parsing algorithm exists. In this work, we review this parsing algorithm and propose important modifications that enlarge the search space. These modifications allow the parsing algorithm to search for more and better solutions. …...

Previous work on dependency parsing used various kinds of combination models but a systematic analysis and comparison of these approaches is lacking. In this paper we imple- mented such a study for English dependency parsing and find several non-obvious facts: (a) the diversity of base parsers is more important than …...

In a supertagging task, sequence labeling models are commonly used. But their lim- ited ability to model long-distance informa- tion presents a bottleneck to make further im- provements. In this paper, we modeled this long-distance information in dependency for- malism and integrated it into the process of HPSG supertagging. The …...

Like most natural language disambiguation tasks, word sense disambiguation (WSD) re- quires world knowledge for accurate predic- tions. Several proxies for this knowledge have been investigated, including labeled cor- pora, user-contributed knowledge, and ma- chine readable dictionaries, but each of these proxies requires significant manual effort to create, and they …...

Open-class semantic lexicon induction is of great interest for current knowledge harvest- ing algorithms. We propose a general frame- work that uses patterns in bootstrapping fash- ion to learn open-class semantic lexicons for different kinds of relations. These patterns re- quire seeds. To estimate the goodness (the po- tential yield) …...

We address the challenge of automatically generating questions from reading materials for educational practice and assessment. Our approach is to overgenerate questions, then rank them. We use manually written rules to perform a sequence of general purpose syn- tactic transformations (e.g., subject-auxiliary inversion) to turn declarative sentences into questions. These …...

Producing a fluent ordering for a set of prenominal modifiers in a noun phrase (NP) is a problematic task for natural lan- guage generation and machine translation systems. We present a novel approach to this issue, adapting multiple sequence alignment techniques used in computa- tional biology to the alignment of …...

This paper describes a method for checking the acceptability of paraphrases in context. We use the Google n-gram data and a CCG parser to certify the paraphrasing grammati- cality and fluency. We collect a corpus of hu- man judgements to evaluate our system. The ultimate goal of our work is …...

We show how features can easily be added to standard generative models for unsuper- vised learning, without requiring complex new training methods. In particular, each component multinomial of a generative model can be turned into a miniature logistic regres- sion model if feature locality permits. The in- tuitive EM algorithm …...

Most existing algorithms for learning latent- variable models—such as EM and existing Gibbs samplers—are token-based, meaning that they update the variables associated with one sentence at a time. The incremental na- ture of these methods makes them suscepti- ble to local optima/slow mixing. In this paper, we introduce a type-based …...

There is significant evidence in the literature that integrating knowledge about multiword expressions can improve shallow parsing ac- curacy. We present an experimental study to quantify this improvement, focusing on com- pound nominals, proper names and adjective- noun constructions. The evaluation set of multiword expressions is derived from Word- Net …...

Adaptor grammars extend probabilistic context-free grammars to define prior dis- tributions over trees with “rich get richer” dynamics. Inference for adaptor grammars seeks to find parse trees for raw text. This paper describes a variational inference al- gorithm for adaptor grammars, providing an alternative to Markov chain Monte Carlo methods. …...

Translation systems are generally trained to optimize BLEU, but many alternative metrics are available. We explore how optimizing toward various automatic evaluation metrics (BLEU, METEOR, NIST, TER) affects the re- sulting model. We train a state-of-the-art MT system using MERT on many parameteriza- tions of each metric and evaluate the …...

State-of-the-art Machine Translation (MT) systems are still far from being perfect. An al- ternative is the so-called Interactive Machine Translation (IMT) framework. In this frame- work, the knowledge of a human translator is combined with a MT system. The vast ma- jority of the existing work on IMT makes use …...

Word Segmentation is the foremost obligatory task in almost all the NLP applications where the initial phase requires tokenization of input into words. Urdu is amongst the Asian languages that face word segmenta- tion challenge. However, unlike other Asian languages, word segmentation in Urdu not only has space omission errors …...

We are interested in diacritizing Semitic lan- guages, especially Syriac, using only dia- critized texts. Previous methods have required the use of tools such as part-of-speech taggers, segmenters, morphological analyzers, and lin- guistic rules to produce state-of-the-art results. We present a low-resource, data-driven, and language-independent approach that uses a hybrid …...

For millions of people in less resourced re- gions of the world, text messages (SMS) pro- vide the only regular contact with their doc- tor. Classifying messages by medical labels supports rapid responses to emergencies, the early identification of epidemics and everyday administration, but challenges include text- brevity, rich morphology, …...