Text-to-Emotion Engine for Real Time Internet Communication

In this paper, the latest version of an emotion extraction engine used for real time Internet text communication is presented. The engine can analyse input text from a chat environment, extract the emotion being communicated, and deliver the parameters necessary to invoke an appropriate expressive image on screen to the communicating user display. The parameters include the emotion extracted from the text being typed and the intensity of the emotion. The measurement of duration of the expression displayed is also possible being an agreed function of the emotion intensity. Semantic analysis is used to extract emotional words. Analysing the individual word position, the person the emotion is referred to, and the time the emotion occurred, identification of emotional words, as well as using of a set of grammatical rules allow the engine to perform satisfactorily. Static and dynamic tests were carried out in order to test the engine performance.