Can we tell which parallel algorithm is executing by looking at the performance of the algorithm? In this work, we design and demonstrate a study for parallel algorithm classification of parallel sorting algorithms. We leverage Caliper to collect the performance data, and Thicket for our exploratory data analysis (EDA). For classification, we develop a workflow with PyTorch and Scikit-learn to evaluate the effectiveness of support vector machines (SVMs), neural networks, and decision trees on parallel performance data. We demonstrate classification accuracy for these three different models of 95.3%, 94.0%, and 90.1% across five different parallel algorithm classes.