| 摘要 |
1. A method of producing a topical dual-language dictionary for use in a topical, dual-language translation system, comprising the steps: inputting a first plurality of references in a first language into said translation system; parsing said first plurality of references into a first plurality of individual words; assigning said first, plurality of parsed words from said first plurality of references to classes; applying said first plurality of parsed words to said topical dual-language dictionary to produce translations of at least some of said first plurality of parsed words; applying any untranslated parsed words from said first plurality of parsed words to a conventional dual-language dictionary to produce translations of said untranslated first plurality of parsed words; inputting a second plurality of references in a second language into said translation system, parsing said second plurality of references into a second plurality of individual words; assigning said second plurality of parsed words from said second plurality of references to classes; applying said second plurality of parsed words to said topical dual-language dictionary to produce translations of at least some of said second plurality of parsed words; applying any untranslated parsed words from said second plurality of parsed words to said conventional dual-language dictionary to produce translations of said untranslated second plurality of parsed words; processing words associations between said first and second pluralities of translated words to produce forward and backward frequency of associations; and updating said topical dictionary based on said frequency of associations. 2. The method of claim 1, wherein inputting a first plurality of references comprises receiving oral voice signals from a telephone, and processing said voice signals using voice recognition circuitry to produce electronic signals representative of said voice signals. 3. The method of claim 1, wherein inputting a first plurality of references comprises: receiving signals from a scanner; and processing said scanner signals using optical character recognition circuitry to convert said scanned signals mio textual information. 4. The method of claim 1, wherein inputting a first plurality of references comprises: receiving textual information via a computer keyboard. 5. The method of claim 1, wherein assigning said first plurality of parsed words comprises: determining the grammatical function of at least a majority of said individual words. 6. The method of claim 1, wherein assigning said first plurality of parsed words comprises: determining the grammatical form of at least a majority of said individual words. 7. The method of claim 1, wherein assigning said first plurality of parsed words comprises: determining the denotation of at least a majority of said individual words. 8. The method of claim 1, wherein assigning said first plurality of parsed words comprises: determining the grammatical function of at least a majority of said individual words; determining the grammatical form of at least a majority of said individual words; and determining the denotation of at least a majority of said individual words. 9. The method of claim 1, further comprising: applying said individual words against a Markov model that associates each individual word against the individual words surrounding each individual word, at least to reduce synonym occurrences. 10. A system for producing a topical dual-language dictionary for use in a topical, dual-language translation system, the system comprising: input circuitry that receives input references; processing circuitry that parses a first plurality of references into a first pluralily of individual words and assigns said first pluralily of parsed words to classes; processing circuitry that applies said first plurality of parsed words to said topical dual-language dictionary to produce translations of at least some of said first plurality of parsed words, and that applies any untranslated parsed words from said first plurality of parsed words to a conventional dual-language dictionary to produce translations of said untranslated first plurality of parsed words; processing circuitry that parses a second plurality of references into a second plurality of individual words and assigns said second plurality of parsed words to classes; processing circuitry that applies said second plurality of parsed words to said topical dual-language dictionary to produce translations of at least some of said second plurality of parsed words, and that applies any untranslated parsed words from said second plurality of parsed words to said conventional dual-language dictionary to produce translations of said untranslated second plurality of parsed words; and processing circuitry that processes word associations between said first and second pluralities of translated words to produce forward and backward frequency of associations, said topical dictionary being updated based on said frequency of associations. 11. The system for producing a topical dictionary of claim 10, wherein said input circuitry comprises a telephone and voice recognition circuitry. 12. The system for producing a topical dictionary of claim 10, wherein said input circuitry comprises a scanner and optical character recognition circuitry. 13. The system for produce a topical dictionary of claim 10, wherein said input circuitry comprises a computer terminal including a video screen and keyboard. 14. The system for producing a topical dictionary of claim 13, wherein said video screen displays said input messages and said computer terminal further comprises means for revising said input message. 15. The system for producing a topical dictionary of claim 13, wherein said computer terminal further comprises means for indicating to an operator that one or more words of said translated message are unclear. |
