Abstract: An automated natural language translation system takes source natural language text (preferably in Japanese) and translates them into a target natural language (preferably English). The system also allows an operator to re-translate automatically selected portions of the source text. The system includes an improvement directed to transforming kanas in the source text into alphabetic letters of the target language which allows the presence of a word or phrase boundary to be recognized in the middle of a kana. The system also includes an improvement involving performing concurrently on the source text both a morphological analysis and a syntactic analysis.
Abstract: A system and method for translating an annotated source document in a first natural language to a target document in a second natural language having corresponding annotations, includes computer storage, a computer receiving module for receiving input textual information in a first language and for storing the input textual information in the computer storage, the input textual information including annotations and a translation engine for creating a first token string including first language tokens, annotation tokens that apply to the first language tokens, and ending tokens. Prior to translation, the annotation tokens are removed from the first token string and a plurality of annotation records are created for the first language tokens, each annotation record linking one of the first language tokens to each of the annotation tokens that apply to the first language token.
Type:
Grant
Filed:
June 14, 1999
Date of Patent:
October 22, 2002
Assignee:
LogoVista Corporation
Inventors:
Lewis G. Pringle, Robert W. Swerdlow, Alec Wysoker
Abstract: Systems and methods for automated natural language translation take source natural language and translate it into a target natural language. The language translation is facilitated through the use of domain specific dictionaries, controlling grammar rules, and probability values associated with some dictionary entries. Domain specific dictionaries provide different translations of words depending on the context in which the words are used. Controlling grammar rules are used to supersede all other grammar rules to eliminate inaccurate and undesirable translations. Probability values are used to determine the likelihood a particular multi-word combination is a particular part of speech.