Deep learning techniques for Sentiment Analysis have become very popular. They provide automatic feature extraction and both richer representation capabilities and better performance than traditional feature based techniques. Traditional surface approaches are based on complex manually extracted features, and this extraction process is a fundamental question in feature driven methods. These long-established approaches can yield strong baselines on their own, and their predictive capabilities can be used in conjunction with the arising deep learning methods.
This master thesis seeks to improve the performance of new deep learning techniques integrating them with traditional surface approaches based on manually extracted features. The contributions of this master thesis are seven-fold. First, we develop a deep learning based sentiment classifier using a word embeddings model and a linear machine learning algorithm. This classifier serves us as a baseline with which we can compare subsequent results. Second, we propose two ensemble techniques which aggregate our baseline classifier with other surface classifiers widely used in the field of Sentiment Analysis. Third, we also propose two models for combining both surface and deep features to merge information from several sources. As fourth contribution, we introduce a taxonomy for lassifying the different models found in the literature, as well as the ones we propose. Fifth, we conduct several experiments to compare the performance of these models with the deep learning baseline. For this, we employ five public datasets that were extracted from the microblogging domain. Sixth, as a result, a statistical study confirms that the performance of these proposed models surpasses that of our original baseline on F1-Score. Finally, an additional case tudy is developed in order to broaden the scope of the proposed models evaluation. This case study introduces two major changes in the evaluation. On the one hand, the domain is different, making the analysis on the review domain. On the other hand, the granularity of the analysis is increased, as it is performed at the aspect based sentiment analysis level.