Algorithm 1. AL experimental protocol.

Inputs$: \mathit{\beta }\to \mathrm{maximum} \mathrm{number} \mathrm{of} \mathrm{iterations}$ $\mathit{s}\to \mathrm{number} \mathrm{of} \mathrm{sampling} \mathrm{examples}$ $\mathit{\theta }\to \mathrm{oracle} \mathrm{for} \mathrm{labelling} \mathrm{unlabelled} \mathrm{examples}$ $\mathit{\gamma }\to \mathrm{multi}-\mathrm{label} \mathrm{AL} \mathrm{strategy}$ $\mathbf{ȹ}\to \mathrm{multi}-\mathrm{label} \mathrm{classification} \mathrm{algorithm}$ $\&{\mathit{T}}_s\to \mathrm{a} \mathrm{test} \mathrm{set} \mathrm{of} \mathrm{multi}-\mathrm{label} \mathrm{examples}$ $\&{\mathit{T}}_r\to \mathrm{a} \mathrm{training} \mathrm{set} \mathrm{of} \mathrm{multi}-\mathrm{label} \mathrm{examples}$

Begin //Construct the labelled and unlabelled sets Ls←Resample⁡(s,Tr); Us←Tr\LS; //Train ȹ with LS ȹ←Trai n(Ls,ȹ); for i ter ←1 to β do   //Select informative example from US   i←SelectInformativeExample (γ,ȹ,US);   //label the selected example   Label(θ,i);   //Update the labelled and unlabelled sets   LS←LS∪i;   US←US\i;   //Train ← with LS   ȹ←Train(Ls,ȹ);   //Evaluate ȹ on TS   Test (Ts,ȹ); end end