A METHOD FOR TRANSLATION OF INFORMATION AND DEVICE FOR THE REALIZATION OF IT

1. A method for machine translation of information given as a character string in a first language into a character string in a second language, comprising - storage in the knowledge base of model segments in the form of character strings in said first language, and in logical connection with these, model segments (133, 134) in the form of character strings in the second language, - identifying a structural segment in the character string of said first language following a first rule (102), - comparing said identified structural segment with model segments (104) in the form of character strings in the first language stored according to a second rule, - striving to select one model segment(110) on the basis of said comparison, - reading a model, i. e. equivalent segment (121) in the form of a character string in the second language logically connected to the selected model segment, and - translating said structural segment into said translation segment in the form of a character string in the second language on the basis of said equivalent segment and a third rule (122), characterized in that the method comprises the identification of an intermediate word and/or a suffix and said stage of identifying a structural segment in the character string is essentially based on the identification of said intermediate word and/or suffix, forming so the beginning and the end of structural segment. 2. A method as claimed in claim 1, characterized in that, said information to be given as a character string in the second language is generated on the basis of translation segments (124). 3. A method as claimed in claim 1 or 2, characterized in that, when no model segment to be selected, in result of comparing structural segments, is found as a result of the comparison of the structural segments, the structural segment is displayed by means of a user interface (131) and the equivalent segment of the displayed structural segment is stored in the knowledge base by means of the user interface (132,133). 4. A method as claimed in any of claims 1-3, characterized in that said structural segment comprises a punctuation mark. 5. A method as claimed in any of claims 1-4, characterized in that the type identifier of the model segment is stored in logical connection with the model segment. 6. A method as claimed in any of claims 1-5, characterized in that there are more than two model segments representing different languages logically connected to each other. 7. A method as claimed i