Evaluating neural word embeddings created from online course reviews for sentiment analysis

摘要: Social media are providing the humus for sharing of knowledge and experiences growth community activities (e.g., debating about different topics). The analysis user-generated content in this area usually relies on Sentiment Analysis. Word embeddings Deep Learning have attracted extensive attention various sentiment detection tasks. In parallel, literature exposed drawbacks traditional approaches when belonging to specific contexts is processed with general techniques. Thus, ad-hoc solutions needed improve effectiveness such systems. paper, we focus coming from e-learning context demonstrate how distributional semantic trained smaller context-specific textual resources more effective respect bigger general-purpose ones. To end, build context-trained online course reviews using state-of-the-art generators. Then, those integrated a deep neural network designed solve polarity task context, modeled as regression. By applying our approach background corpora contexts, show that performance better aligned regression context.