a distributed attribute reduction algorithm based on cosine similarity under the Spark framework Please, follow the following instructions to use DARCS.
To be able to use the DARCS code, you will need to install the following:
Install Scala 2.11.12
Install Spark 2.4.7
rawDataRDD = The input data set
selectedAttributesIndex = The index of attributes in the candidate attribute subset
nbIterIfPerFeat = Number of iterations
sizeColumns = The number of attributes in each partitioned data block