Table 5 The classifier \(p (\mathbf {C} |\mathbf {Q})\) with equivalent to Cosine similarity between the trained model \(p(\mathbf {C}|\mathbf {T})\) for each class and new samples \(\mathbf {Q} = \{Q_1,\dots ,Q_6 \}\) as a query. \(\mathbf {C}=\{C_1,C_2,C_3 \}\) is three states of the fan. The similarity results in the prediction for the new samples. The maximum value of the cosine similarity for each sample is boldfaced

From: Fog computing application of cyber-physical models of IoT devices with symbolic approximation algorithms