Login
Computer Science @ Rochester
Friday, February 03, 2012
12:45 PM
CSB 703
Darcey Riley
University of Rochester
Confused-Based Statistical Language Modeling
NLP systems frequently use discriminative language models to rerank their outputs. These language models can help to correct for common errors that are made by the system. Each system will make different types of errors; for instance, the incorrect outputs of a speech recognition system might contain homonyms such as "their" instead of "there", and these would be very different from the sorts of errors made by a machine translation system. Discriminative models are particularly well-suited to the task of reranking because they can explicitly model incorrect data and thus learn to avoid system-specific mistakes.

Unfortunately, discriminative models require a large amount of supervised training data, which may not be available when working with resource-poor languages. The training data for discriminative rerankers usually consists of actual outputs from the system as well as correct outputs provided by a human annotator. For machine translation, for instance, the reranker would be trained on a parallel corpus: a French sentence would be translated into a number of possible English sentences, which would be compared against the gold standard translation.

Confusion-based language modeling is a semisupervised method for training the reranker using only a plain monolingual corpus. In order to do this, we must simulate possible "confusions", incorrect sentences that the system would be likely to confuse with the correct output. The CLSP held a workshop on this topic last summer; this talk will describe the research that was done there.