Algorithm 1. Pig wasting disease sound classification procedure.

Input: a matrix of training samples A = [A1, A2,…, Ac] ∈ ℝm×n for c classes, a test sample y ∈ ℝm. Normalize the columns of A to have unit l2 norm. Solve the l1 minimization problem: ${\widehat{\alpha }}_1=\mathit{\text{arg}}\underset{\alpha }{min}{\Vert \alpha \Vert }_1\text{subject to}y=A\alpha .$ Compute the mean coefficient value of each class. $m{c}_{\text{i}}=\frac{1}{n_i}\text{∑}_1^{n_i}{\delta }_i({\widehat{\alpha }}_1)\text{for}i=1,2,\dots ,c$ mci: mean coefficient value of i – th class; ni: number of elements in i – th class; δi(α̂1): characteristic function that selects the coefficients associated with the i – th class; Sort the class in descending order according with mean coefficient value of classes. Output: pig wasting diseases, which have large mean coefficient value.