Possible Usage of Sentiment Analysis for Calculating Vectors of Felific Calculus

In this paper we introduce an algorithm for affective reasoning based on Bent ham´s Felific Calculus known also as the hedonic calculus. Knowledge recquired for the task is retrived from a blog corpus by means of sentiment analysis on sentences containing an action or state input. This approach allows a machine to gather information on how usually other people feel when something happens, why people did it and what could happen after the act. Such knowledge is important for understanding actions of others, and for acquiring emphatic skills by a machine. In addition to emotion categorization of Nakamura, we introduce two lexicons based on McDougall´s instinct classification and Kohlbergian stages of moral development, then show some basic efficiency of the retrieved knowledge.