Year: 2007
Type: conference-paper
Abstract: In this paper, we introduce context informed features in a log-linear phrase-based SMT framework; these features enable us to exploit source similarity in addition to target similarity modeled by the language model. We\npresent a memory-based classification framework that enables the estimation of these features while avoiding\nsparseness problems. We evaluate the performance of our approach on Italian-to-English and Chinese-to-English translation tasks using a state-of-the-art phrase-based SMT\nsystem, and report significant improvements for both BLEU and NIST scores when adding the context-informed features.
Authors: Nicolas Stroppa, Antal van den Bosch, Andy Way
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1568694948
