In this paper, we describe a two-stage multilingual dependency parser used for the multilingual track of the CoNLL 2007 shared task. The system consists of two components: an unla...
Trigram language models are compressed using a Golomb coding method inspired by the original Unix spell program. Compression methods trade off space, time and accuracy (loss). The...
We describe some challenges of adaptation in the 2007 CoNLL Shared Task on Domain Adaptation. Our error analysis for this task suggests that a primary source of error is differenc...
Mark Dredze, John Blitzer, Partha Pratim Talukdar,...
We present a two-stage multilingual dependency parsing system submitted to the Multilingual Track of CoNLL-2007. The parser first identifies dependencies using a deterministic p...
This paper presents a large-scale system for the recognition and semantic disambiguation of named entities based on information extracted from a large encyclopedic collection and ...