People have multiple accounts on Online Social Networks (OSNs) for various purposes. It is of great interest for third parties to collect more users' information by linking their accounts on different OSNs. Unfortunately, most users have not been aware of potential risks of such accounts linkage. Therefore, the design of a control methodology that allows users to share their information without the risk of being linked becomes an urgent need, yet still remains open.In this paper, we first aim to raise the users' awareness by presenting an effective User Accounts Linkage Inference (UALI), which is shown to be more powerful to users than existing methods. In order to help users control the risks of UALI, we next propose the first Information Control Mechanism (ICM), in which users' information is still visible as intended and, in the meanwhile, the risk of their accounts linkage can be controlled. Using real-world datasets, the performance of ICM is validated, and we also show that it works well for various linkage inference approaches. Both UALI and ICM approaches, designed to take generic inputs, extend their ability to be widely applied into many practical social services.