Jimmy Callin: The Distribution of Mood An Exploration of Distributional Compositions in Sentiment Classification

On this day, Jimmy Callin defended his Bachelor’s thesis in language technology in Uppsala. He has investigated how compositionality could be implemented on the level of distributional vectors. The results show that operations beyond simple vector addition probably will be necessary to be able to do something interesting with this sort of model: the simplest additive operations did not beat a simple tfidf model.