The Mathematics of Statistical Machine Translation: Parameter Estimation

1 Introduction

1. What does Brown et al. (1993) describe?
   A series of five statistical translation models (the IBM Models 1–5) with EM algorithms for estimating their parameters from pairs of sentences that are translations of each other, plus word-by-word alignments; trained on the French–English Canadian Hansards.
   • Brown et al., 1993, The Mathematics of Statistical Machine Translation (Computational Linguistics), §1
2. What lesson did sentence-alignment work (Brown, Lai & Mercer; Gale & Church) teach, which this paper extends?
   Simple statistical methods can be surprisingly successful: sentences could be aligned using only their lengths (in words or characters), without looking at the words. This paper goes on to align the words within aligned sentence pairs.
   • Brown et al., 1993, The Mathematics of Statistical Machine Translation (Computational Linguistics), §1
3. Why did the authors believe word alignments are “inherent” in a large bilingual corpus?
   Their algorithms have minimal linguistic content (the only linguistic input was rules for finding sentence boundaries), yet they recover convincing alignments, so the alignments must come from the data itself.
   • Brown et al., 1993, The Mathematics of Statistical Machine Translation (Computational Linguistics), §1