AirlDM

AirlDM is a collection of machine learning algorithms, which are data source independent through the means of sufficient statistics and data source wrappers. They work with general data sources (where data can be stored in any format) as long as wrappers for accessing and getting sufficient statistics from those data sources are provided. Some of the algorithms in AirlDM are adapted from Weka implementations by separating the statistics gathering and hypothesis generation components. The figure below shows the general architecture of AirlDM.