Open-domain dialogue generation: Presence, limitation and future directions

Dialogue systems have received more and more attention from researchers, especially open-domain dialogue generation system. The application scenario of it, also chat-bot, is very extensive, from website customer service to dialogue robot such as Siri and Microsoft Xiaoice. More and more studies have been done on open-domain dialogue generation. In recent years, deep learning is gradually emerging which can leverage a massive amount of data to learn meaningful feature representations. In this paper, we first give the development of deep learning in Natural Language Processing(NLP) especially in dialogue generation. Next we analyze the current problems and introduce several improvements which are the current research hotspot. The purpose of this paper is to sort out the research results of recent years and make contributions to the research and development of the dialogue systemin thefuture.