Abstract: Smart environment research has resulted in many useful tools for modeling, monitoring, and adapting to a single resident. However, many of these tools are not equipped for coping with multiple residents in the same environment simultaneously. In this paper we investigate a first step in coping with multiple residents, that of attributing sensor events to individuals in a multi-resident environment. We discuss approaches that can be used to achieve this goal and we evaluate our implementations in the context of two physical smart environment testbeds. We also explore how learning resident identifiers can aid in performing other analyses on smart environment sensor data such as activity recognition.