Unsupervised Lemmatization as Embeddings-Based Word Clustering

TL;DR

This paper introduces an unsupervised lemmatization method using agglomerative clustering with a novel distance measure combining string similarity and word embeddings, effective across multiple languages.

Contribution

It proposes a new unsupervised clustering approach for lemmatization that leverages combined string and embedding similarities, outperforming baselines in diverse languages.

Findings

Outperforms baseline on 23 of 28 datasets

Effective across 23 languages

Combines string similarity with embedding cosine similarity

Abstract

We focus on the task of unsupervised lemmatization, i.e. grouping together inflected forms of one word under one label (a lemma) without the use of annotated training data. We propose to perform agglomerative clustering of word forms with a novel distance measure. Our distance measure is based on the observation that inflections of the same word tend to be similar both string-wise and in meaning. We therefore combine word embedding cosine similarity, serving as a proxy to the meaning similarity, with Jaro-Winkler edit distance. Our experiments on 23 languages show our approach to be promising, surpassing the baseline on 23 of the 28 evaluation datasets.