## Qualitative and Quantitative Models Of Speech Translation (1994)

Citations: | 8 - 1 self |

### BibTeX

@MISC{Alshawi94qualitativeand,

author = {Hiyan Alshawi},

title = {Qualitative and Quantitative Models Of Speech Translation},

year = {1994}

}

### OpenURL

### Abstract

This paper compares a qualitative reasoning model of translation with a quantitative statistical model. We consider these models within the context of two hypothetical speech translation systems, starting with a logic-based design and pointing out which of its characteristics are best preserved or eliminated in moving to the second, quantitative design. The quantitative language and translation models are based on relations between lexical heads of phrases. Statistical parameters for structural dependency, lexical transfer, nd linear order are used to select a set of implicit relations between words in a source utterance, a corresponding set of relations between target language words, and the most likely translation of the original utterance.

### Citations

608 | Interpretation as abduction
- Hobbs, Stickel, et al.
- 1993
(Show Context)
Citation Context ... that constrain the word sense predicates in all contexts. A is a set of assumptions sufficient to support the interpretation φ given S 7and R. In other words, this is ‘interpretation as abduction’ (=-=Hobbs et al. 1988-=-), since abduction, not deduction, is needed to arrive at the assumptions A. The most common types of meaning postulates in R are those for restriction, hyponymy, and disjointness, expressed as follow... |

565 | Distributional clustering of English words - Pereira, Tishby, et al. - 1993 |

471 | Generalized Phrase Structure Grammar - Gazdar, Klein, et al. - 1984 |

444 | An introduction to unificationbased approaches to grammar
- Shieber
- 1986
(Show Context)
Citation Context ...ral language. The qualitative-quantitative distinction can also be seen as underlying the difference between classification systems based on feature specifications, as used in unification formalisms (=-=Shieber 1986-=-), and clustering based on a variable degree of granularity (e.g. Pereira, Tishby and Lee 1993). It seems unlikely that these continuously variable aspects of fluent natural language can be captured b... |

307 | Structural ambiguity and lexical relations - Hindle, Rooth - 1993 |

282 |
Neural network classifiers estimate Bayesian a posterior probabilities
- Richard, Lippmann
- 1991
(Show Context)
Citation Context ... and Carter 1994). Another example is connection strengths in neural network approaches to language processing, though it has been shown that certain networks are effectively computing probabilities (=-=Richard and Lippmann 1991-=-). Nevertheless, probability theory does offer a coherent and relatively well understood framework for selecting between uncertain alternatives, making it a natural choice for quantitative language pr... |

222 | English Word Grammar
- Hudson
- 1990
(Show Context)
Citation Context ...ure. 9Instead of feature based syntax trees and first-order logical forms we will adopt a simpler, monostratal representation that is more closely related to those found in dependency grammars (e.g. =-=Hudson 1984-=-). Dependency representations have been used in large scale qualitative machine translation systems, notably by McCord (1988). The notion of a lexical ‘head’ of a phrase is central to these representa... |

143 | Word sense disambiguation using a second language monolingual corpus - Dagan, Itai - 1994 |

133 | Automatic grammar induction and parsing free text: A transformation-based approach
- Brill
- 1993
(Show Context)
Citation Context ...ified according to its performance against corpora of natural text (e.g. Taylor, Grover, and Briscoe 1989). Second, there is a class of techniques for learning rules from text, a recent example being =-=Brill 1993-=-. Conversely, it is possible to imagine building a language model in which all probabilities are estimated according to intuition without reference to any real data, giving a probabilistic model that ... |

115 | Monotonic semantic interpretation
- Alshawi, Crouch
- 1992
(Show Context)
Citation Context ... intended by the speaker. Although we have not provided a denotational semantics for sets of relation edges, we anticipate that this will be possible along the lines developed in monotonic semantics (=-=Alshawi and Crouch 1992-=-). 6.2 Translation Parameters To be practical, a model for P(Ct|Cs) needs to decompose the source and target graphs Cs and Ct into subgraphs small enough that subgraph translation parameters can be es... |

83 | Contextual word similarity and estimation from sparse data. Computer Speech and Language - Dagan, Marcus, et al. - 1995 |

72 | Automatically extracting and representing collocations for language generation - Smadja, MeKeown - 1990 |

63 |
rinciples of lexical language modeling for speech recognion
- Jelinek, Mercer, et al.
- 1991
(Show Context)
Citation Context ... communication, a connection that has been exploited in speech recognition, for example using the concept of entropy to provide a motivated way 3of measuring the complexity of a recognition problem (=-=Jelinek et al. 1992-=-). Even if probability theory remains, as it currently is, the method of choice in making language processing quantitative, this still leaves the field wide open in terms of carving up language proces... |

59 | Training and scaling preference functions for disambiguation
- Alshawi, Carter
- 1994
(Show Context)
Citation Context ...hich the trees conform to, say, minimal attachment and parallelism between conjuncts. Such functions have been used in tandem with statistical functions in experiments on disambiguation (for instance =-=Alshawi and Carter 1994-=-). Another example is connection strengths in neural network approaches to language processing, though it has been shown that certain networks are effectively computing probabilities (Richard and Lipp... |

29 | GPSM: a Generalized Probabilistic Semantic Model for ambiguity resolution - Chang, Luo, et al. - 1992 |

28 |
A Statistical Approach to Machine Translation, Computational Linguistics
- Brown, Cocke, et al.
- 1990
(Show Context)
Citation Context ...nts for probability theory to work with. For translation, a very direct approach using parameters based on surface positions of words in source and target sentences was adopted in the Candide system (=-=Brown et al. 1990-=-). However, this does not capture important structural properties of natural language. Nor does it take into account generalizations about translation that are independent of the exact word order in s... |

17 | The syntactic regularity of English noun phrases - Taylor, Grover, et al. - 1989 |

16 | A Corpus-Based Statistics-Oriented Transfer and Generation Model for Machine Translation
- Chang, Su
- 1993
(Show Context)
Citation Context ...t language string that is the most likely translation of the input. We are thus interested in the conditional probability of Wt given As. This conditional probability can be expressed as follows (cf. =-=Chang and Su 1993-=-): P(Wt|As) = ∑ Ws,Cs,Ct P(Ws|As) P(Cs|Ws,As) P(Ct|Cs,Ws,As) P(Wt|Ct,Cs,Ws,As). We now apply some simplifying independence assumptions concerning relation graphs. Specifically, that their derivation f... |

13 | Deriving Database Queries from Logical Forms by Abductive Definition Expansion - Rayner, Alshawi - 1992 |

9 |
SwedishEnglish QLF Translation
- Alshawi, Carter, et al.
- 1992
(Show Context)
Citation Context ...dependent of the exact word order in source and target sentences. Such generalizations are, of course, central to qualitative structural approaches to translation (e.g. Isabelle and Macklovitch 1986, =-=Alshawi et al. 1992-=-). The aim of the quantitative language and translation models presented in sections 5 and 6 is to employ probabilistic parameters that reflect linguistic structure without discarding rich lexical inf... |

6 | A Multi-Target Machine Translation System - McCord - 1988 |

5 | Roossin: A Statistical Approach to Machine Transla tion - Brown, John, et al. - 1990 |

3 |
Implementing Systemic Classification by Unification”. Computational Linguistics 14:40–51
- Mellish
- 1988
(Show Context)
Citation Context ...r restriction, hyponymy, and disjointness, expressed as follows: p1(x1,x2) → p2(x1) restriction; p2(x) → p3(x) hyponymy; ¬(p3(x) ∧ p4(x)) disjointness. Although there are compilation techniques (e.g. =-=Mellish 1988-=-) which allow selectional constraints stated in this fashion to be implemented efficiently, the scheme is problematic in other respects. To start with, the assumption of a small set of senses for a wo... |

1 | Training and Sealing Preference Functions for Disambiguation - Alshawi, Carter - 1994 |

1 | Word Grammar - Itudson - 1984 |

1 | Transfer and - Isah, P, et al. - 1986 |

1 | Principles of Lexieal Language Modeling for Speech Recognition - Iclinek, Mercer, et al. - 1992 |

1 | Implementing Systemic Classification by Unification - llish, S - 1988 |

1 | Information Based ,';yntax and Semantics: Volume I ~ Fundamentals - I'olltrd, Sag - 1987 |

1 | Neural Network Classiliers Estimate Bayesian a posteriori Prob;d,ilili~.s". Neural Computation - iichard, Lippmann - 1991 |

1 | An Introduction to UnificationIlascd Approaches to Grammar. CSLI Lecture Notes, Number 4. Center for the Study of Language and I i~ form ation - Shi - 1986 |

1 |
Information Based Syntax and Semantics: Volume I — Fundamentals
- Pollard, Sag
- 1987
(Show Context)
Citation Context ...he phrase may be combined with other phrases. This notion has been central to a number of approaches to grammar for some time, including theories like dependency grammar (Hudson 1976, 1990) and HPSG (=-=Pollard and Sag 1987-=-). More recently, the statistical properties of associations between words, and more particularly heads of phrases, has become an active area of research (e.g. Chang, Luo, and Su 1992; Hindle and Root... |