Morfessor EM+Prune: Improved Subword Segmentation with Expectation Maximization and Pruning
This work addresses subword segmentation for NLP applications like speech recognition and machine translation, offering incremental improvements to existing methods.
The paper tackled the problem of subword segmentation for natural language processing by proposing improved training algorithms based on Expectation Maximization and pruning, showing that this approach finds better solutions than the original Morfessor Baseline model and leads to higher morphological segmentation accuracy on English, Finnish, North Sami, and Turkish datasets.
Data-driven segmentation of words into subword units has been used in various natural language processing applications such as automatic speech recognition and statistical machine translation for almost 20 years. Recently it has became more widely adopted, as models based on deep neural networks often benefit from subword units even for morphologically simpler languages. In this paper, we discuss and compare training algorithms for a unigram subword model, based on the Expectation Maximization algorithm and lexicon pruning. Using English, Finnish, North Sami, and Turkish data sets, we show that this approach is able to find better solutions to the optimization problem defined by the Morfessor Baseline model than its original recursive training algorithm. The improved optimization also leads to higher morphological segmentation accuracy when compared to a linguistic gold standard. We publish implementations of the new algorithms in the widely-used Morfessor software package.