Abstract
The research of crowdsourcing developed rapidly in recent years. We can accomplish some request in a short time by people on Internet. Also we can ask friends to accomplish the requests, which is called friendsourcing. It is important to find out the ways to increase accomplish rate of a request. In this research, I focus on the crowdsourcing article on unstructured crowdsourcing platform. I use the Chinese version of Linguistic Inquiry and Word Count(LIWC) to analyze the difference of the linguistic pattern between crowdsourcing article and non-crowdsourcing article. I figure out the difference of the crowdsourcing articles written by a same user on different platform. I use the result to build the crowdsourcing article generating assistance system. I also find out the design implication of this system through interview and user testing. This system can help people to write proper crowdsourcing article on a specific crowdsourcing platform.