Maybe other scientists know tools for supervised learning, like PSO.
But if you could use alternatively a tool for unsupervised learning, we are currently working on a Toolbox in R for swarm based clustering and visualization of high dimensional data, see
Article Self Organized Swarms for cluster preserving Projections of ...
You may be searching for the Rough sets theory (RST). really the classical RST don't depend on search techniques but recently many researches were done to use the search techniques for finding the best minimal subset before classifying the data using the rough set.
You may want to try WEKA. I'm not sure if there is any Swarm intelligent-based classifiers in it, I know it has GA and many other classifiers and also clustering algorithms.
All swarm intelligent algorithms (like ACO, PSO, Bees algorithms, fish algorithm, monkey algorithm, cat algorithm, conile algorithm, creative algorithm, and so on so far) can be used as a helpful tools for the classifier which I mean to help the classifier to get best classification accuracy, but if you want to use swarm intelligent algorithms as a classifier, there is only ant minor algorithm work as a classifier and it is based on ACO which is consider as a one of swarm intelligent algorithm.
I want to know what features is input into ACO algorithm? I mean that other papers explain that they use features which are meta tag, url, tagging term and bag of term. I want to use bag of words to use in classification. So I want to need deeply explanation ACO algorithm for text classification step by step or example if u have reference. I search google and other site, they give me TSP problem so I don't clearly understand. Help me please. Thank you all..
Y. del Valle, G. Venayagamoorthy, S. Mohagheghi, J.-C. Hernandez, R. Harley, Particle swarm optimization: Basic concepts, variants and applications in power systems, IEEE Transactions on Evolutionary Computation, 12 (2) (2008) 171-195.