In this session, we will be taking a closer look at one of the most popular semantic techniques currently in
use, Sentiment Analysis. This is most often a supervised Machine Learning task which uses pure word extraction, term weighting, and classification to predict whether a text document is
positive or negative. However, as we see in some of this session's readings, there has also been research into using unsupervised Machine Learning, as well as taking into account various aspects of language - such as parts-of-speech and semantic frames.