The task of variable selection is essential to improving the ability of machine learning systems to generalize. Although there are many conventional variable selection methods, almost all of them need to prepare and learn a large number of samples in advance because they are based on offline learning. This property is not suitable for online learning systems.To overcome this inconvenience, we propose a quick online variable selection method inspired by human problem solving behaviors. The proposed method tries to generate several variable set candidates in a speculative manner using a filter method and evaluates them using a wrapper method. The method can also function in concept-drifting environments, where relevant variable sets are changing. The experimental results show that the new method yields appropriate variable sets from a small number of samples.at each step and yields the best one as the solution. The method can yield the appropriate solution quickly and can adapt to changing environments, so-called concept-drifting environments ¡ 1-4244-0991-8/07/$25.00/©2007 IEEE