Unsupervised Morphological Expansion of Small Datasets for Improving Word Embeddings
This addresses the problem of limited training data for word embeddings in multiple languages, though it appears incremental as it builds on existing embedding techniques.
The paper tackles data sparsity in word embeddings by using an unsupervised morphological expansion method to generate artificial sentences, improving results across all seven languages tested, with specific gains on eleven test sets for word similarity.
We present a language independent, unsupervised method for building word embeddings using morphological expansion of text. Our model handles the problem of data sparsity and yields improved word embeddings by relying on training word embeddings on artificially generated sentences. We evaluate our method using small sized training sets on eleven test sets for the word similarity task across seven languages. Further, for English, we evaluated the impacts of our approach using a large training set on three standard test sets. Our method improved results across all languages.