Sentiment Analysis of Iraqi Arabic Dialect on Facebook Based on Distributed Representations of Documents

摘要: Nowadays, social media is used by many people to express their opinions about a variety of topics. Opinion Mining or Sentiment Analysis techniques extract from user generated contents. Over the years, multitude studies has been done English language with deficiencies research in all other languages. Unfortunately, Arabic one languages that seems lack substantial research, despite rapid growth its use on outlets. Furthermore, specific dialects should be studied, not just Modern Standard Arabic. In this paper, we experiment sentiments analysis Iraqi dialect using word embedding. First, made large corpus previous works learn representations. Second, embedding model training Doc2Vec representations based Paragraph and Distributed Memory Model Vectors (DM-PV) architecture. Lastly, represented feature for four binary classifiers (Logistic Regression, Decision Tree, Support Vector Machine Naive Bayes) detect sentiment. We also experimented different values parameters (window size, dimension negative samples). light experiments, it can concluded our approach achieves better performance Logistic Regression than classifiers.