Amazon Technologies, Inc.

William Joseph Kaper

Kelly Duggar Wiggins

Thibault Pierre Seillier

Paul Christopher Cerda

Paul Kasper

Pallami Bhattacharjee

Gavin Richard Jewell

Gregory Alan Hanneman

Alon Lavie

Michael Denkowski

Alexandre Alexandrovich Klementiev

Robert Thomas Olszewski

Matthew Ryan Fiorillo

Jeremy Boynes

Addison Phillips

Christopher A Heilman

David Armour

William Cary Hall

Robert E Hicks

Austin Matthews

Christopher James Dyer

Kartik Goyal

Ehud Hershkovich

Laura Josephine Kieras

A content management system (CMS) and a translation management system (TMS) can utilize content dimensions for content items to manage and translate the content items between languages. Machine and human translations of complex dynamic content can also be improved by pre-rendering the content to remove localization-related syntax prior to machine or human translation. Content items can also be scored as to their suitability for localization prior to translation, and translation can be skipped for content items that do not have a sufficiently high score. Semantic and natural language processing (NLP) techniques can also be utilized for content categorization and routing. Translations of content items can also be continuously refined and higher quality re-translated content can be provided in an automated fashion.

Source language content scoring for localizability

Technologies are disclosed herein for statistical machine translation. In particular, the disclosed technologies include extensions to conventional machine translation pipelines: the use of multiple domain-specific and non-domain-specific dynamic language translation models and language models; cluster-based language models; and large-scale discriminative training. Incremental update technologies are also disclosed for use in updating a machine translation system in four areas: word alignment; translation modeling; language modeling; and parameter estimation. A mechanism is also disclosed for training and utilizing a runtime machine translation quality classifier for estimating the quality of machine translations without the benefit of reference translations. The runtime machine translation quality classifier is generated in a manner to offset imbalances in the number of training instances in various classes, and to assign a greater penalty to the misclassification of lower-quality translations as higher-quality translations than to misclassification of higher-quality translations as lower-quality translations.

Optimized statistical machine translation system with rapid adaptation capability

Semantic and natural language processing for content categorization and routing

A system for conducting a code review of program source code including translation of content in the program source code may include a source control system. The system may also include a resource file translator to facilitate translation of the content from a first language into second languages. Upon receipt of a code review request, the source control system may notify a workflow manager that the resource files contain the content for translation, and the workflow manager may notify a resource file translator that the program source code contains the content for translation. The resource file translator may cause translation of the resource files into translated resource files. The source control system may receive program source code under review from a code reviewer interface, and the resource file translator may update the program source code under review with the translated resource files.

Code review with automated translation

Leveraging content dimensions during the translation of human-readable languages

Abstraction of syntax in localization through pre-rendering

Growing community of inventors

Pittsburg, PA, United States of America

William Joseph Kaper