A Projection Extension Algorithm for Statistical Machine Translation

A Projection Extension Algorithm for Statistical Machine Translation Christoph Tillmann IBM T.J. Watson Research Center Yorktown Heights, NY 10598 ctill@us.ibm.com Abstract In this paper, we describe a phrase-based unigram model for statistical machine translation that uses a much simpler set of model parameters than similar phrase- based models. The units of translation are blocks – pairs of phrases. During decod- ing, we use a block unigram model and a word-based trigram language model. Dur- ing training, the blocks are learned from source interval projections using an un- derlying high-precision word alignment. The system performance is significantly increased by applying a novel block exten- sion algorithm using an additional high- recall word alignment. The blocks are fur- ther filtered using unigram-count selection criteria. The system has been successfully test on a Chinese-English and an Arabic- English translation task. 1 Introduction Various papers use phrase-based translation systems (Och et al., 1999; Marcu and Wong, 2002; Ya- mada and Knight, 2002) that have shown to improve translation quality over single-word based transla- tion systems introduced in (Brown et al., 1993). In this paper, we present a similar system with a much simpler set of model parameters. Specifically, we compute the probability of a block sequence �� � . A block � is a pair consisting of a contiguous source and a contiguous target phrase. The block sequence Figure 1: A block sequence that jointly generates � target and source phrases. The example is actual decoder output and the English translation is slightly incorrect. probability ��� � � � is decomposed into conditional probabilities using the chain rule: � �� � � �  �   � ��� �  �  � (1)  �   �  � �  "! �  $# � �   �  � We try to find the block sequence that maximizes ��� �%� � : �%� � '&(*) +,&�-/.*0 1 � �� �%� �$ . The model pro- posed is a joint model as in (Marcu and Wong, 2002), since target and source phrases are generated jointly. The approach is illustrated in Figure 1. The source phrases are given on the 2 -axis and the target phrases are given on the 3 -axis. During block decod- ing a bijection between source and target phrases is generated. The two types of parameters in Eq 1 are defined as: 4 Block unigram model 576*8 9;: : We compute un- igram probabilities for the blocks. The blocks are simpler than the alignment templates (Och et al., 1999) in that they do not have an internal structure. 4 Trigram language model: the probability 576*8<9>= 8%9?A@ : between adjacent blocks is com- puted as the probability of the first target word in the target clump of 8 9 given the final two words of the target clump of 8 9?A@ . The exponent B is set in informal experiments to be CEDGF . No other parameters such as distortion proba- bilities are used. To select blocks …

Content truncated to 500 words