Who said chatbots were cold-hearted?
A deep learning-based chatbot capable of assessing the emotional content of a conversation and responding accordingly. Emotional Chatting Machine results from Chinese researchers’ work on the design of a new generation of emotionally aware chatbots. To do so, the researchers trained the algorithm on a dataset of 23,000 sentences collected from the Chinese blogging service Weibo and manually annotated with their emotional charge – anger, disgust, happiness, like, sadness. The chatbot can then generate appropriate responses in terms of both content and emotion.
The ability to perceive and express emotions is a step towards the long-term goal of artificial intelligence: a machine that understands affect and emotion (cf. Oshi-el). But here turning emotions into data is inevitably reductive and remains based on a human ranking of just five reactions.
Spotted in 2018