Evaluation of different methods for Arabic WordNet Expansion and sentiment lexicon development.

What is it about?

In this article, we investigate multiple link prediction approaches and different natural language processing variations in order to try to link a large lexical Arabic Resource, SAMA, to English WordNet. We use a machine learning approach and we develop a training data using existing small scale Arabic WordNet. We use the gold data set to evaluate the different link prediction techniques and we perform a detailed error analysis for each of the techniques. Finally, we use the best approach to create a large-scale Arabic sentiment lexicon, ArSenL 2.0, using SAMA and English SentiWordNet.

Featured Image

Why is it important?

The article is important because it provides publicly a large-scale sentiment lexical resource for Arabic, ArSenL 2.0, to enable more accurate sentiment mining models. It also presents a benchmark dataset to be used by other researchers in the field interested in automatically evaluating approaches for Arabic WordNet Expansion. Although current state-of-the-art models in NLP, namely deep learning models, are based on raw data, integrating accurate lexical resources would definitely improve the accuracy of the models.