Журнал дистанционного зондирования и ГИС

Абстрактный

Companionbot

Suranjana Trivedy

The aim of the project is to build an intelligent naturally dialog able machine, which can talk with the depressed and lonely older people and young people and finally become their true companion. Specifically, the research goal was to build an automatic system to understand multimodal emotion, including facial expressions, speech tone and the linguistic emotions. For that purpose author used semi-supervised learning methods, Natural language processing techniques (NLP) and Information retrieval (IR) Term frequency – Inverse document frequency (TF-IDF) techniques to analyze emotion from text. She has extracted emotion from facial images. Overall aim is to create intelligent companionbot. Here they can see how machine can act in a different way with older and young people.