LU-BZU at SemEval-2021 Task 2: Word2Vec and Lemma2Vec performance in Arabic Word-in-Context disambiguation

TL;DR

This study evaluates the effectiveness of Word2Vec and Lemma2Vec models in Arabic Word-in-Context disambiguation, comparing their performance without relying on sense inventories, using datasets from SemEval-2021.

Contribution

It introduces a comparative analysis of Word2Vec and Lemma2Vec models for Arabic WiC disambiguation without sense inventories, highlighting their relative performance.

Findings

Lemma2Vec models outperform Word2Vec in disambiguation accuracy

Pre-trained models on large Arabic corpora improve results

Lemma-based models show advantages over word-based models in this task

Abstract

This paper presents a set of experiments to evaluate and compare between the performance of using CBOW Word2Vec and Lemma2Vec models for Arabic Word-in-Context (WiC) disambiguation without using sense inventories or sense embeddings. As part of the SemEval-2021 Shared Task 2 on WiC disambiguation, we used the dev.ar-ar dataset (2k sentence pairs) to decide whether two words in a given sentence pair carry the same meaning. We used two Word2Vec models: Wiki-CBOW, a pre-trained model on Arabic Wikipedia, and another model we trained on large Arabic corpora of about 3 billion tokens. Two Lemma2Vec models was also constructed based on the two Word2Vec models. Each of the four models was then used in the WiC disambiguation task, and then evaluated on the SemEval-2021 test.ar-ar dataset. At the end, we reported the performance of different models and compared between using lemma-based and word-based models.