Russian word sense induction by clustering averaged word embeddings

TL;DR

This paper presents a simple clustering-based approach using averaged word embeddings for Russian word sense induction, demonstrating competitive results and highlighting the effectiveness of small, balanced corpora for training embeddings.

Contribution

Introduces a naive clustering method with pre-trained embeddings for Russian word sense induction, showing small, balanced corpora can outperform larger noisy datasets.

Findings

Achieved 2nd place on wiki-wiki dataset

Small, balanced corpora can outperform large noisy data in sense induction

Simple averaging and clustering can be effective for word sense tasks

Abstract

The paper reports our participation in the shared task on word sense induction and disambiguation for the Russian language (RUSSE-2018). Our team was ranked 2nd for the wiki-wiki dataset (containing mostly homonyms) and 5th for the bts-rnc and active-dict datasets (containing mostly polysemous words) among all 19 participants. The method we employed was extremely naive. It implied representing contexts of ambiguous words as averaged word embedding vectors, using off-the-shelf pre-trained distributional models. Then, these vector representations were clustered with mainstream clustering techniques, thus producing the groups corresponding to the ambiguous word senses. As a side result, we show that word embedding models trained on small but balanced corpora can be superior to those trained on large but noisy data - not only in intrinsic evaluation, but also in downstream tasks like word sense induction.