A Replication Study of the Top Performing Systems in SemEval Twitter Sentiment Analysis

We performed a thorough replicate study of the top performing systems in the yearly SemEval Twitter Sentiment Analysis task. We highlight and discuss differences among the results obtained by those systems that have been officially published and the ones we are able to compute. Learning from the studies being made on the systems, we also propose SentiME, an ensemble system composed of five state-of-the-art sentiment classifiers. SentiME trains the different classifiers using the Bootstrap Aggregating Algorithm. The classification results are then aggregated using a linear function that averages the classification distributions of the different classifiers. SentiME has also been evaluated over the SemEval2015 test set, properly trained with the SemEval2015 train test. We show that SentiME would outperform the best ranked system of the challenge.