Evaluation Of Word Embeddings From Large-Scale French Web Content

TL;DR

This paper introduces and evaluates high-quality French word embeddings trained on large-scale web data, demonstrating their effectiveness across NLP tasks and providing accessible tools for further use.

Contribution

It presents new French word vectors trained on massive web data, evaluates their quality on analogy and NLP tasks, and offers a demo application and open resources.

Findings

Pretrained embeddings outperform existing French vectors.

Embeddings significantly improve NLP task performance.

Open resources facilitate further research and application.

Abstract

Distributed word representations are popularly used in many tasks in natural language processing. Adding that pretrained word vectors on huge text corpus achieved high performance in many different NLP tasks. This paper introduces multiple high-quality word vectors for the French language where two of them are trained on massive crawled French data during this study and the others are trained on an already existing French corpus. We also evaluate the quality of our proposed word vectors and the existing French word vectors on the French word analogy task. In addition, we do the evaluation on multiple real NLP tasks that shows the important performance enhancement of the pre-trained word vectors compared to the existing and random ones. Finally, we created a demo web application to test and visualize the obtained word embeddings. The produced French word embeddings are available to the public, along with the finetuning code on the NLU tasks and the demo code.