Task adaptation using MAP estimation in N-gram language modeling

This paper describes a method of task adaptation in N-gram language modeling, for accurately estimating the N-gram statistics from the small amount of data of the target task. Assuming a task-independent N-gram to be a-priori knowledge, the N-gram is adapted to a target task by MAP (maximum a-posteriori probability) estimation. Experimental results showed that the perplexities of the task adapted models were 15% (trigram), 24% (bigram) lower than those of the task-independent model, and that the perplexity reduction of the adaptation went up to 39 % at maximum when the amount of text data in the adapted task was very small.