Support Vector Machines (SVMs) are powerful classification and regression algorithms that work on a wide variety of problem sets. This effort looks at how SVMs perform on a multi-dimensional, multi-class classification problem of human activities based on imprecise radio-frequency identifier (RFID) sensors. This analysis provides the background of the data, explores the algorithm and its specific application, provides the results, and discusses the findings.