Abbasi, Ahmed, Hsinchun Chen, and Arab Salem. "Sentiment analysis in multiple languages: Feature selection for opinion classification in web forums." ACM transactions on information systems (TOIS) 26.3 (2008): 1-34.

This study proposes using sentiment analysis methodologies to classify Web forum opinions in multiple languages, focusing on English and Arabic content. It evaluates the utility of stylistic and syntactic features for sentiment classification, incorporating specific feature extraction components for Arabic linguistic characteristics. The study introduces the entropy weighted genetic algorithm (EWGA), a hybridized genetic algorithm that uses the information-gain heuristic for feature selection to improve performance and identify key features. The proposed features and techniques are tested on a benchmark movie review dataset and U.S. and Middle Eastern Web forum postings. Results show that using EWGA with SVM achieves high accuracy levels, exceeding 91% across all testbeds. Stylistic features notably enhance performance, and EWGA outperforms other feature selection methods, demonstrating the effectiveness of these approaches for document-level sentiment classification.