ArSenL: A Large Scale Arabic Sentiment Lexicon

What is it about?

ArSenL is a publicly available lemma-based Arabic sentiment lexicon that was automatically annotated for three sentiment scores positive, negative and neutral using English SentiwordNet. It helps to improve the accuracy of opinion mining models from the text.

Featured Image

Why is it important?

Using sentiment lexicons in sentiment analysis showed to be successful in enhancing the accuracy of such classification systems for English. While several resources exist for English, only a few exist for Arabic and are publicly available. ArSenL is large-scale, publicly available and lemma-based which simplifies its integration in any Arabic NLP related task.

Perspectives