A study of the use of multi‐objective evolutionary algorithms to learn Boolean queries: A comparative study
Antonio Gabriel López-Herrera, Enrique Herrera‐Viedma, Francisco Herrera · Journal of the American Society for Information Science and Technology · 2009
Abstract In this article, our interest is focused on the automatic learning of Boolean queries in information retrieval systems (IRSs) by means of multi‐objective evolutionary algorithms considering the classic performance criteria, precision and recall. We present a comparative study of four well‐known, general‐purpose, multi‐objective evolutionary algorithms to learn Boolean queries in IRSs. These evolutionary algorithms are the Nondominated Sorting Genetic Algorithm (NSGA‐II), the first version of the Strength Pareto Evolutionary Algorithm (SPEA), the second version of SPEA (SPEA2), and the Multi‐Objective Genetic Algorithm (MOGA).