The search engine, which is very helpful for information searching, is essential to the present Internet. However, its powerful features lead to some problems, e.g., too many search results or they are spread over various documents. These situations take a lot of time for data filtering. There have been the relative studies using Semantic Web to resolve these problems. With the large growth of News contents, the system will consume much time to retrieval News Information in some researches related to the topic of News System. In this study, we want to solve this problem by using Cloud Computing and implement a Semantic-Based inference function in a News Query System. The system can be used not only to enhance the recall ratio but also to reduce the processing time for the information retrieval.