Apparatus and method for automatic translation customized for documents in restrictive domain

摘要

An apparatus and method for automatic translation of patent documents are provided. Translation knowledge customized for the patent domain is extracted and used. The method includes the steps of: extracting a number of technical terms from a patent document and assigning translated terms for the extracted technical terms; extracting expressions frequently used in patent documents, not in a document in a general domain, and constructing translated expressions; using the translation knowledge constructed in this way, applying a sentence pattern to extraordinarily long sentences in a patent document, dividing the sentence by recognition of parallel structure and syntax clues; and extracting translation units available for parsing and performing structure analysis.