Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Division
Browse By Session Type
Search Tips
Personal Schedule
Sign In
In this paper we use several supervised machine learning approaches and compare their success in predicting the sentiment of Austrian (German language) parliamentary speeches and news reports. Prediction results in learning-based sentiment analysis vary strongly. They depend on the choice of algorithm and its parameterization, the quality and quantity of available training data as well as the selection of appropriate input feature representations. Our training data contain human-annotated sentiment scores at the phrase and sentence level. Going beyond the dominant bag-of-words modeling approach in traditional natural language processing, we also test sentiment analysis for neural network-based distributed representations of words. The latter reflect syntactic as well as semantic relatedness, but require huge amounts of training examples. We test both approaches with heterogeneous textual data, compare their success rates and provide conclusions on how to improve the sentiment analysis of political communication.
Elena Sofie Rudkowsky, U of Vienna
Martin Haselmayer
Matthias Wastian, Technical U Vienna
Marcelo Jenny, U of Vienna
Stefan Emrich, Drahtwarenhandlung Vienna
Michael Sedlmair, U of Vienna