This article presents an online cluster using genetic algorithms to increase information retrieval efficiency. The Information Retrieval (IR) is based on the grouping of documents. Documents with high similarity to group are judge more relevant to the query and should be retrieved more efficiently. Under genetic algorithms, an individual is a hierarchical chromosome with all the documents of a documental base; and we generate a population of different individuals. These chromosomes feed into genetic operator process: selection, crossover, and mutation until we get an optimize cluster chromosome for document retrieval. Our testing result show that information retrieval with 0.9 crossover probability and 0.65 mutation probability give the highest precision while lower crossover probability and high mutation probability give the highest recall. © 2008 IADIS.