Two FPGA Devices in the Problem of Finding Minimal Reducts
Mateusz Choromański , Tomasz Grześ , Piotr Hońko
AbstractSpeeding up attribute reduction process is an important issue in data mining. The goal of this paper is to compare two hardware implementations of minimal reducts computation, i.e. the previously introduced implementation on Intel Arria V SoC and a newly proposed solution on Xilinx Zynq Ultrascale+ MPSoC. Two versions of an attribute reduction algorithm, i.e. blind and frequency based breadth search strategies, were implemented on the two frameworks. Experimental research showed that finding minimal reducts can be accelerated several times when using the new device.
|Publication size in sheets||0.5|
|Book||Saeed Khalid, Rituparna Chaki, Janev Valentina (eds.): Computer Information Systems and Industrial Management : 18th International Conference : CISIM 2019 : proceedings, Lecture Notes In Computer Science, no. 11703, 2019, Springer, ISBN 978-3-030-28956-0, 540 p., DOI:10.1007/978-3-030-28957-7|
|Keywords in English||Attribute reduction Minimal reducts Breadth search strategy Hardware implementation Field programmable gate arrays|
|Internal identifier||ROC 19-20|
|Score||= 40.0, 30-03-2020, ChapterFromConference|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.