Mnogoznal: An unsupervised system for word sense disambiguation

In this paper, we present Mnogoznal, an unsupervised system for word sense disambiguation. Given a sentence, the system chooses the most relevant sense of each input word w.r.t. to the semantic similarity between the given sentence and the synset constituting the sense of the target word. Mnogoznal has two modes of operation. The sparse mode uses the traditional vector space model to estimate the most similar word sense corresponding to its context. The dense mode, instead, uses synset embeddings to cope with the sparsity problem. We describe the architecture of the present system and also conduct its preliminary evaluation on three different lexical semantic resources for Russian. We found that the dense mode substantially outperform the sparse one on all the datasets as according to the adjusted Rand index computed on a gold standard dataset.