Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... boundary, neural network, multilingual, corpus 1 Introduction An important step in many multilingual text processing tasks, including sentence alignment (Kay Roscheisen 93; Gale Church 93) automatic lexicon construction (Melamed 95) and knowledge base acquisition for machine translation (Mitamura et al. 93) is the segmentation of texts into individual sentences. The sentence boundary recognition system SATZ, Palmer Hearst 94) Palmer 94) provides an efficient, trainable algorithm for this task. The approach in the SATZ system is to represent the context surrounding a punctuation mark as a ....

Teruko Mitamura, Eric H. Nyberg, and Jaime G. Carbonell. Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for mt. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan, 1993.

....time, we also carried out a bottom up development strategy. Recognizing the importance of the corporate multilingual history captured by the significant corpus of already translated documentation, we were able to leverage this resource in order to determine the lexical inventory of the domain (see [Mitamura et al. 1993]) A wide range of automated natural language processing tools including text deformatters, KWIC editors, multilingual Source Interlingua French Spanish X . Analysis French Generation x Generation Spanish Generation Figure 1: The Interlingual MT Architecture corpus aligners, and ....

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.

.... interacts with a lexical disambiguator during processing (see section 6) The lexicon for the heavy equipment domain contains approximately 54,000 nomenclature phrases which, like the single word vocabulary, have been extracted from a large cross section of the domain using corpus analysis [Mitamura et al. 1993]. These phrases each have a single sense in the lexicon. The author is encouraged to always select the most specific phrase available in the lexicon for the meaning that he or she wishes to convey. Thus, the author may choose e.g. either of the terms floor mat or rubber mat when the mat in the ....

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and MethodologicalIssues in Machine Translation, Kyoto, Japan.

No context found.

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.

....During the development of KANT applications, we have tried whenever possible to make use of on line resources (dictionary, corpora, etc. and analysis techniques (indexing, alignment, extraction, etc. to support the process of acquiring domain and linguistic knowledge for an application (Mitamura, Nyberg, and Carbonell, 1993; Leavitt et al. 1994) Given the apparent success of the CATALYST system, we can say that KBMT systems can indeed be scaled up, and that corpus analysis and on line resources are certainly useful. But for highly accurate technical translation, complete automation does not seem to be a feasible ....

....for their participation in the project. Notes 1. For example, extraction of English French term pairs from translated texts could only account for 12 of the initial CATALYST French lexicon development, because the small number of aligned texts that were available did not cover the entire domain (Mitamura, Nyberg, and Carbonell, 1993). 2. Concept names are prefixed by the gross semantic category class specifiers O , A , P , M , and U to represent objects, actions, properties, manners, and units respectively. 3. For the sake of expedience, simple heuristics were utilized for constructing likely noun phrases based on the ....

Mitamura, T., E. Nyberg, and J. Carbonell. 1993. Automated Corpus Analysis and the Acquisition of Large, Multi-Lingual Knowledge Bases for MT. In Proceedings of TMI-93.

No context found.

Mitamura, Nyberg and Carbonell (1993). "Automated Corpus Analysis and the Acquisition of Large, Multi-Lingual Knowledge Bases for MT", Proceedings of TMI-93.

No context found.

Mitamura, T, E. Nyberg and J. Carbonell (1993). "Automated Corpus Analysis and the Acquisition of Large, Multi-Lingual Knowledge Bases for MT," Proceedings of TMI-93.

....for that phrase [3] To achieve these benefits, however, there is an associated cost, which in this case is the burden of constructing the source and target language lexica on a large scale. The construction of the source language lexicon is largely automated from analysis of existing corpora [5], but the construction of the generation lexicon requires more effort, both because its format includes more explicit syntactic information (i.e. a full f structure for each target language term) and because online corpora are often not available in the target language. It is the acquisition of ....

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.

....important to reduce the processing overhead associated withambiguityresolutionin order to keep the system fast enough for on line use. 2.1 The Domain Lexicon The domain lexicon is built using corpus analysis. Lists of terms, arranged by part of speech, are automatically extracted from the corpus [Mitamura et al. 1993] . The lexicon consists of closed class general words, open class general words, idioms, and nomenclature phrases. Closed class general words (e.g. the, with, should) are taken from general English. Open class general words (e.g. drain, run, hot) are limited in the lexicon to one sense per part of ....

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.

No context found.

Mitamura, T., Nyberg, E., and Carbonell, J. (1993). Automated corpus analysis and the acquisition of large, multi-lingual knowledge bases for MT. In 5th International Conference on Theoretical and Methodological Issues in Machine Translation, Kyoto, Japan.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC