Documentation » Historique » Révision 17
Révision 16 (François Rioult, 17/01/2011 13:06) → Révision 17/37 (François Rioult, 31/01/2011 18:40)
h1. Documentation KDAriane is a set of operators for data mining and machine learning, and a set of scenarios (supervised classification, missing values completion, strong emerging pattern mining, etc.). It uses Ariane as a graphical platform for designing the data streams. h2. Installation * [[Prerequisite]] * [[KDAriane]] h2. How is Ariane working? Ariane is a graphical platform Special operators for designing image processing streams. Ariane works with graphical operators, and allows to build loops and while. "Go to the Aiane dedicated site":http://www.greyc.ensicaen.fr/~regis/Ariane/ for a full documentation about Ariane. shell scripting h2. How is KDAriane working KDAriane is provided with basic components for executing shell scripts. The choice depends on how many parameters (p), input (i) and output (o) you want. The operators are named @"eval" + p + i + o @ and call the eponymous .sh script. When an operator is executed, Ariane launches the script (for example @script.sh@) associated to the operator with giving the following arguments: <pre> script.sh parameter-1 parameter-2 ... parameter-p input-1 input-2 ... input-i output-1 output-2 ... output-o </pre> In Ariane, every operator has a return value, even if it has no output. See [Special operators for shell scripting] for more details. The operator are divided in two categories: * [[KDD operators]] are special components for calling Weka components or RapidMiner processes. * [[Shell operators]] that directly execute the commands entered by Ariane. h2. Pattern mining prototypes * [[music-dfs]] : mining patterns under various constraints * [[mtminer]] : levelwise minimal transversals of hypergraph h2. Scenarios KDAriane provides some examples of KDD realized through Ariane: * [[Data preparation]] : a first scenario for the binarization of CSV data. * pattern mining and complexity visualization * supervised classification with association rules * experiences about perturbation on training and test file with Weka classifiers and RapidMiner processes.