Columbus provides a declarative framework of operations for feature selection over in-RDBMS data. The framework is integrated into the R environment, and is usable as a library in R.
What Columbus Is:Columbus is a scalable framework to perform exploratory feature selection. Columbus system is implemented using R as the front end and oracle at the back end.
How Columbus Works:Columbus system supports a common set of feature selection operations as primitives, that can be used by analyst to perform feature selection dialogue. By being aware of the computational properties of the operations, Columbus system optimizes the performance by applying ideas like batching and materialization. Moreover, Columbus system tracks the intermediate steps and results there by enabling the system to answer the queries posed by the analyst and provide better insight into the feature selection dialogue.
The source code and datasets are coming soon!
Support and Collaboration
Columbus is kindly supported by Oracle. We also thank the generous support given by the National Science Foundation CAREER Award under IIS-1054009 and the Office of Naval Research under award no. N000141210041.