Emotion Analysis of the Text Using Fuzzy Affect Typing over Emotions

This paper presents a novel approach of emotion estimation in the affective content of the textual messages or dialogues. The individual words in a sentence has been chopped and mapped onto the corresponding affective categories. An affective category is assigned to its membership value in the all basic 5 emotions viz. happy, surprise, neutral, anger and sad. To analyze the affective content effectively we use natural language processing for the lexical analysis and thereafter fuzzy affect typing over basic emotions with the membership modifier rules to handle various modifiers of affective contents. Machine learning based prediction approach has been suggested for new encountered words.