Sparsity in Social Robotics Experiments

The paper addresses the use of several performance indexes, namely the Gini sparsity index (GI) and the Bhattacharyya coefficient (BC), to analyze data acquired in contexts such as social robotics (SR) experiments. The claim is that using only the indexes common in SR can be misleading and data sparsity can reveal the presence of abnormalities. These ideas are illustrated with experiments with synthetic data.