Abstract: The status of the network media trend is introduced and a new system “Semantic Analysis Interactive Knowledge System” (SAIKS) is presented in this paper. By applying natural language understanding technology, the system searches a relevant answer in the knowledge database intelligently, and then chooses an optimized answer, which helps the users to find the answers to their questions conveniently, rapidly and accurately. At the same time, the knowledge database is improved by users' feedback. This paper studies the system key modules, introduces the system architecture in detail, constructs the experiment system, and gives out preliminary results.