LearningWord Embeddings for Low-resource Languages by PU Learning
This addresses the challenge of building effective word embeddings for low-resource languages, which is incremental as it adapts existing PU-learning methods to a specific bottleneck in NLP.
The paper tackles the problem of learning word embeddings for low-resource languages with limited text data by proposing a Positive-Unlabeled Learning approach to factorize sparse co-occurrence matrices, achieving improved performance in experiments across four languages.
Word embedding is a key component in many downstream applications in processing natural languages. Existing approaches often assume the existence of a large collection of text for learning effective word embedding. However, such a corpus may not be available for some low-resource languages. In this paper, we study how to effectively learn a word embedding model on a corpus with only a few million tokens. In such a situation, the co-occurrence matrix is sparse as the co-occurrences of many word pairs are unobserved. In contrast to existing approaches often only sample a few unobserved word pairs as negative samples, we argue that the zero entries in the co-occurrence matrix also provide valuable information. We then design a Positive-Unlabeled Learning (PU-Learning) approach to factorize the co-occurrence matrix and validate the proposed approaches in four different languages.