Datasets used for the following SNAM paper: --------------------------------------------------------------------------------------------------- Title: From Classification to Quantification in Tweet Sentiment Analysis Authors: Wei Gao and Fabrizio Sebastiani Organization: Qatar Computing Research Institute, Hamad Bin Khalifa University, Doha, Qatar --------------------------------------------------------------------------------------------------- [Content] * SemEval2013, SemEval2014, SemEval2015 datasets: - semeval.train.feature.txt: Training set for learning sentiment models at development stage - semeval.dev.feature.txt: Held-out set for tuning parameters - semeval.train+dev.feature.txt: Training set for learning the final sentiment model - semeval13.test.feature.txt: SemEval2013 test set - semeval14.test.feature.txt: SemEval2014 test set - semeval15.test.feature.txt: SemEval2015 test set * Other datasets: semeval2016, sanders, sst, omd, hcr, gasp, wa, wb - X.train.feature.txt: Training set for learning sentiment models at development stage - X.dev.feature.txt: Held-out set for tuning parameters - X.train+dev.feature.txt: Training set for learning the final sentiment model - X.test.feature.txt (or X.dev-test.feature.txt for semeval2016 only): Test set where X is one of semeval2016, sanders, sst, omd, hcr and gasp. * Training files are saved in ./data/train directory, and held-out and test files are in ./data/test directory For more details, please refer to the paper. [Citation] You can cite the following paper when referring to the dataset: @article{gao2016classification, title={From classification to quantification in tweet sentiment analysis}, author={Gao, Wei and Sebastiani, Fabrizio}, journal={Social Network Analysis and Mining}, volume={6}, number={1}, pages={19}, year={2016}, publisher={Springer} }