US 6,275,789 B1Granted

Method and apparatus for performing full bidirectional translation between a source language and a linked alternative language

OwnerLeo Moser, Robert Moserinventors

Life of this patent

Estimating the term…

Loading patent…

Abstract

System with apparatus to improve international and other communication, and to provide easier access to data, especially digitized data, by means of linked alternative languages generated from a source language. As taught by the present invention, a linked alternative language is an especially designed language form quite different in outward format from its source language in that it has been optimized in a plurality of ways to allow targeted populations to comprehend and use it more efficiently than the source language, but which has also been carefully designed to retain full bidirectional machine translation equivalence to the source language. All use of artificial intelligence and computational linguistics for machine translation as taught in the present invention is constrained by these considerations.

Claims21

Dependent claims

An apparatus for providing bidirectional translations of a text of data between a linked alternative language and a source language, wherein the source language is a natural language and the linked alternative language is designed to map fully to the source language in terms of structure and strings of digitized data, comprising:

(a) means for entering the text of data;
(b) a dictionary database of a vocabulary of words in the source language stored as records, with collocated information on the usage pattern of each word and on frequency of use of each word in the source language;
(c) a dictionary database for a vocabulary of words in the linked alternative language stored as records;
(d) means for storing the dictionary database of the vocabulary of words in the source language and the dictionary database of the vocabulary of words in the linked alternative language in a central concordance, the records within the dictionary databases taking the form of strings of digitized data in the linked alternative language and in the source language;
(e) a database of instructions which index relationships between the strings of digitized data in the linked alternative language and the strings of digitized data in the source language;
(f) a database of translation rules, wherein all of the translation rules in the database of translation rules provide a lossless translation between the linked alternative language and the source language wherein the linked alternative language is designed to map fully to the source language in terms of structure and strings of digitized data;
(g) automated means for translating, in both directions, between the linked alternative language and the source language, wherein the means for translating applies to the text the set of translation rules and the two dictionary databases stored within the central concordance;
and
(h) means for outputting translated text.

6 more elements

A method for creating and employing a linked alternative language, wherein the linked alternative language is linked to a source language, and is designed to map fully to the source language in terms of structure and strings of digitized data, and wherein the linked alternative language is further designed to provide communicative features and efficiencies on the computer beyond those provided by the source language, the method comprising the steps of:

(a) establishing the parameters of the system, comprising the steps of:
(1) choosing the source language;
(2) targeting the user group which the linked alternative language is to serve;
and
(3) choosing the communicative features and the efficiencies on computer systems to be accommodated by the linked alternative language;
(b) entering into a computer a dictionary database of vocabulary in the source language, with collocated information on the usage pattern of each word and on its frequency of use in the source language;
(c) entering a framework for a dictionary database for the vocabulary of the linked alternative language, the framework being structured to map to the dictionary database of vocabulary in the source language;
(d) building the lexical records within the dictionary database for the vocabulary of the linked alternative language, comprising the steps of:
(1) supplying the linked alternative language with a phonetic system generally reflecting the speech habits of its targeted speakers;
(2) creating a graphemic system to provide a method for writing the linked alternative language in a manner which reflects its phonetic system and is compatible with computer capabilities;
(3) using a computer to screen the morphemes tentatively chosen for the linked alternative language to assure that no two morphemes are so close phonetically as to lead to a serious confusion among the targeted speakers;
and
(4) supplying the linked alternative language with a system for establishing sentence structure which is capable of retaining full computer-implemented mapping to the source language;
(e) storing the two dictionary databases within the computer in a central concordance, the records within the dictionary databases taking the form of strings of digitized data in the linked alternative language and in the source language;
(f) entering a database of instructions which index the relationships between the strings of digitized data in the linked alternative language and the strings of digitized data in the source language;
(g) implementing, on a computer, a set of translation rules wherein there is fully accurate automated and lossless translation in both directions between the linked alternative language and the source language;
and
(h) outputting translated text.

15 more elements

A method implemented on a computer for translating in both directions between a linked alternative language and a source language, wherein the linked alternative language is designed to map fully to the source language in terms of structure and strings of digitized data, comprising the steps of:

(a) entering a text of data into the computer system;
(b) dividing the text of data into sentences;
(c) consulting a central concordance, wherein the central concordance contains strings of digitized data, including spaces and punctuation marks, in the linked alternative language and in the source language, and instructions on relationships between the strings of digitized data;
(d) identifying within each sentence in the text of data those strings of digitized data, which appear in the central concordance, wherein the concordance contains strings of digitized data in the linked alternative language and the source language, and instructions on relationships between the strings of digitized data entering a database of instructions which index the relationships between the strings of digitized data in the linked alternative language and the strings of digitized data in the source language;
(e) implementing, on the computer, a set of translation rules wherein each of the translation rules allows for a filly accurate automated and lossless translation in both directions between the linked alternative language and the source language to the text, wherein the translation rules determine whether identified strings of digitized data should be substituted with related strings of digitized data in the concordance and wherein the translation rules establish the order in which the strings of digitized data are to appear in the translated text;
and
(f) outputting translated text.

4 more elements

Dependencies are inferred from the claim text.