Robust Measurement and Comparison of Context Similarity for Finding Translation Pairs

In cross-language information retrieval it is often important to align words that are similar in meaning in two corpora written in different languages. Previous research shows that using context similarity to align words is helpful when no dictionary entry is available. We suggest a new method which selects a subset of words (pivot words) associated with a query and then matches these words across languages. To detect word associations, we demonstrate that a new Bayesian method for estimating Point-wise Mutual Information provides improved accuracy. In the second step, matching is done in a novel way that calculates the chance of an accidental overlap of pivot words using the hypergeometric distribution. We implemented a wide variety of previously suggested methods. Testing in two conditions, a small comparable corpora pair and a large but unrelated corpora pair, both written in disparate languages, we show that our approach consistently outperforms the other systems.