Microblogging Opinion Mining Approach for Kuwaiti Dialect

In this paper, we present an approach to extract and classify opinion in microblogging. It is based mainly on linguistic resources produced for the Kuwaiti dialect used by an SVM classifier. The approach has been tested on a corpus of 340,000 tweets about "interrogation of ministers by the National Assembly of Kuwait -??????? ??????? " during the last two years. Tweets were collected automatically by a module developed in java. This corpus has been manually annotated by three Kuwaiti dialect native speakers. We obtained an average value of precision and recall respectively 76% and 61%.