Improved statistical alignment models

TLDR

The paper reviews IBM alignment models, a Hidden‑Markov model, smoothing techniques, and various modifications. The paper presents and compares single‑word based alignment models for statistical machine translation. We present and compare various single‑word based alignment models, propose methods to combine alignments, and evaluate them by comparing the resulting Viterbi alignment quality to a manually produced reference alignment. Models with first‑order dependence and a fertility model significantly outperform simple models IBM‑1 and IBM‑2, which cannot capture beyond zero‑order dependencies.

Abstract

In this paper, we present and compare various single-word based alignment models for statistical machine translation. We discuss the five IBM alignment models, the Hidden-Markov alignment model, smoothing techniques and various modifications. We present different methods to combine alignments. As evaluation criterion we use the quality of the resulting Viterbi alignment compared to a manually produced reference alignment. We show that models with a first-order dependence and a fertility model lead to significantly better results than the simple models IBM-1 or IBM-2, which are not able to go beyond zero-order dependencies.