This study proposes a word alignment model based on a recurrent neural network (RNN), in which an unlimited alignment history is represented by recurrently connected hidden layers.

Our alignment model is directional, similar to the generative IBM models (Brown et al., 1993).

the HMM alignment model and achieved state-of-the-art performance.

We assume that this property would fit with a word alignment task, and we propose an RNN-based word alignment model .

The NN-based alignment models are supervised models.

Various word alignment models have been proposed.

2.1 Generative Alignment Model

2.2 FFNN-based Alignment Model

A first-order HMM alignment model (Vogel et al., 1996) is an HMM of length I + 1 where the hidden state at position i E [I ]0 is the aligned index j E [J ]0, and the transition score takes into account the previously aligned index 3" E [J ]0.1 Formally, define the set of possible HMM alignments as X C {0,1}([I]0X[J]0)U([I]X[J]0X[J]0) with

The directional bias of the e—>f and f —>e alignment models may cause them to produce differing alignments.

In this work, we instead consider a bidirectional alignment model that jointly considers both directional models.

We have introduced a novel Lagrangian relaxation algorithm for a bidirectional alignment model that uses incremental constraint addition and coarse-to-fine pruning to find exact solutions.

Our experimental results compare the accuracy and optimality of our decoding algorithm to directional alignment models and previous work on this bidirectional model.

They employed a word alignment model to capture opinion relations among words, and then used a random walking algorithm to extract opinion targets.

They have investigated a series of techniques to enhance opinion relations identification performance, such as nearest neighbor rules (Liu et al., 2005), syntactic patterns (Zhang et al., 2010; Popescu and Etzioni, 2005), word alignment models (Liu et al., 2012; Liu et al., 2013b; Liu et al., 2013a), etc.

(Liu et al., 2012; Liu et al., 2013a; Liu et al., 2013b) employed word alignment model to capture opinion relations rather than syntactic parsing.

Thus assuming there is an alignment model that is able to tell how likely one relation maps to the original question, we add extra alignment-based features for the incoming and outgoing relation of each node.

We describe such an alignment model in § 5.

Since the relations on one side of these pairs are not natural sentences, we ran the most simple IBM alignment Model 1 (Brown et al., 1993) to estimate the translation probability with GIZA++ (Och and Ney, 2003).