A new method to create a sentiment lexicon for the Arabic language using semi-supervised learning on the WordNet and matching them with an Arabic database is presented.
Sentiment Analysis is the process of identifying sentiment from text written in a natural language with respect to the entity it is referring to. Sentiment lexicons are used to perform this task. Several lexicons are available to perform this task in English using WordNet. In this paper, we present a new method to create a sentiment lexicon for the Arabic language using semi-supervised learning on the WordNet and matching them with an Arabic database.