Regularized F-Measure Maximization for Feature Selection and Classification

Algorithm 1

𝐿₂ regularized F-measure maximization.

1. Given 𝛾, 𝜆, a small number 𝜀, Initialize 𝐰^𝑡=𝐰⁰, and set

  𝑡=0.

2. While |𝐰^𝑡+1−𝐰^𝑡|>𝜀

  𝐰^𝑡+1=𝐰^𝑡+𝜇(∇𝐸₂), where 𝜇 is the step-size

3. 𝑡=𝑡+1