Please allow me to introduce Apache Joshua (Incubating). Joshua has a passion for languages... a serious passion. Although absolutely brand new to the Apache Incubator, Joshua needs little introduction within the Statistical Machine Translation (STM) arena. This presentation promises to deliver on Joshua's motivation, functional capabilities and roadmap as the community progress down the path to graduation. In particular this presentation will cover some of the many features offered by Joshua including (i) support for both phrase-based and syntax-based decoding models (ii) translation of weighted input lattices (iii) Thrax: a Hadoop-based, scalable grammar extractor, and (iv) Joshua's sparse feature architecture. The presentation details ongoing use of Joshua within a number of DARPA projects which aim to advance online search capabilities far beyond the current state of the art.
