In a data streaming model, a data owner releases records or documents to a set of users with matching interests, in such a way that the match in interest can be calculated from the correlation between each pair of document and user query. For scalability and availability reasons, this calculation is delegated to third-party servers, which gives rise to the need to protect the integrity and privacy of the documents and user queries. In this paper, we propose a server-aided data stream monitoring scheme (DSM) to address the aforementioned integrity and privacy challenges, so that the users are able to verify the correlation scores obtained from the server. The scheme provides strong security protection, even in the event of collusion between the server and other users. We also offer techniques to bound the computation demand in decoding the correlation scores, and we demonstrate the practicality of the scheme through experiments with real data.