Technologies for statistical machine translation based on generated reordering knowledge

A statistical machine translation (SMT) system comprising:

• a computer;
• a parsing module implemented by the computer and configured to receive a word string in a first natural language and parse said word string into a parse tree comprising a plurality of child nodes, the parse tree representing a syntactic structure of the word string;
• a preprocessing module implemented by the computer and coupled with said parsing module, said preprocessing module configured to access said plurality of child nodes and reorder words from said word string based on a syntax of said first natural language and on a plurality of alignment matrices that map first sample sentences in the first natural language with second sample sentences in a second natural language to provide a plurality of reordered word strings;
• and
• a decoding module implemented by the computer and coupled with said preprocessing module, said decoding module configured to access said plurality of reordered word strings, identify a statistically preferred reordered word string based on reordering probabilities associated with said plurality of reordered word strings, and generate a target word string based on a word sequence of said statistically preferred reordered word string.

A language reordering system for use in statistical machine translation (SMT), said language reordering system comprising:

• a computer;
• a training database for storing training data comprising sentences in a first natural language paired with sentences in a second natural language;
• an alignment model implemented by the computer and configured to match words and phrases in said first natural language to words and phrases in said second natural language, said alignment model utilizing said training data to generate training samples identifying syntactic differences between said first natural language and said second natural language;
• and
• a preprocessing module implemented by the computer and coupled with said training database and said alignment model, said preprocessing module configured to generate a body of reordering knowledge based on a syntax of said first natural language and on a plurality of alignment matrices that map first sample sentences in the first natural language with second sample sentences in the second natural language.