MapReduce accelerated attribute reduction based on neighborhood entropy with Apache Spark

作者：

Highlights：

• Horizontal partitioning method for neighborhood-entropy computation is proposed.

• Conceptual parallelization logic of attribute reduction is presented.

• MapReduce accelerated attribute reduction algorithm is implemented using Spark.

• Experiments confirm the scalability and extensibility of the proposed algorithm.

摘要

•Horizontal partitioning method for neighborhood-entropy computation is proposed.•Conceptual parallelization logic of attribute reduction is presented.•MapReduce accelerated attribute reduction algorithm is implemented using Spark.•Experiments confirm the scalability and extensibility of the proposed algorithm.

论文关键词：Attribute reduction,Neighborhood rough sets,Uncertainty measure,Parallel computing,Apache Spark

论文评审过程：Received 19 March 2022, Revised 22 July 2022, Accepted 12 August 2022, Available online 18 August 2022, Version of Record 26 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118554