Human computer conversation is regarded as one of the most difficult problemsin artificial intelligence. In this paper, we address one of its keysub-problems, referred to as short text conversation, in which given a messagefrom human, the computer returns a reasonable response to the message. Weleverage the vast amount of short conversation data available on social mediato study the issue. We propose formalizing short text conversation as a searchproblem at the first step, and employing state-of-the-art information retrieval(IR) techniques to carry out the task. We investigate the significance as wellas the limitation of the IR approach. Our experiments demonstrate that theretrieval-based model can make the system behave rather "intelligently", whencombined with a huge repository of conversation data from social media.
展开▼