Textual Affect Sensing for Computational Advertising

In this paper, a technique for the generation of advertising headlines based on affect is described. The system takes as input the target concept to evoke and possibly a sentiment to communicate. A fundamental dimension of advertising relies on the affective connotation of words. We have used it as an important component of our algorithms for the creative variation of familiar expressions such as proverbs and famous movie titles. Recognition of affective connotation is based on a semantic similarity acquired in an unsupervised way from large corpora and it allows the system selecting words that are slanted towards a desired emotion, or producing strong ironic effects.