LSTM Network for Inflected Abbreviation Expansion

In this paper, the problem of recovery of morphological information lost in abbreviated forms is addressed with a focus on highly inflected languages. Evidence is presented that the correct inflected form of an expanded abbreviation can in many cases be deduced solely from morphosyntactic tags of the context. The prediction model is a deep bidirectional LSTM network with tag embedding. The network is trained on over 10 million words from the Polish Sejm Corpus and achieves 74.2% prediction accuracy on a smaller, but more general National Corpus of Polish. Analysis of errors suggests that performance in this task may improve if some prior knowledge about the abbreviated word is incorporated into the model.