A novel approach based on genetic algorithm to speed up the discovery of classification rules on GPUs
作者：
Highlights：
• A new approach to produce classification rules based on evolutionary computation is proposed.
• Two novel concepts; coverage matrix and reduction vectors and an altered form of the reduction vector is proposed.
• Novel crossover and mutation operators customized for execution on graphics processing unit (GPU) is proposed.
• The maximum accuracy achieved is 99.74%, 95.73% and 100% for Hepatitis C Virus (HCV), Poker and COVID-19 datasets, respectively.
• The maximum speedup achieved is 23.06% for HCV, 22.12% for COVID-19, and 57.15% for Poker, compared to using single core processors.
摘要
•A new approach to produce classification rules based on evolutionary computation is proposed.•Two novel concepts; coverage matrix and reduction vectors and an altered form of the reduction vector is proposed.•Novel crossover and mutation operators customized for execution on graphics processing unit (GPU) is proposed.•The maximum accuracy achieved is 99.74%, 95.73% and 100% for Hepatitis C Virus (HCV), Poker and COVID-19 datasets, respectively.•The maximum speedup achieved is 23.06% for HCV, 22.12% for COVID-19, and 57.15% for Poker, compared to using single core processors.
论文关键词：Data mining,Machine learning,Rule discovery,Genetic algorithm,GPU programming,Classification rules
论文评审过程：Received 1 April 2020, Revised 17 August 2021, Accepted 18 August 2021, Available online 21 August 2021, Version of Record 3 September 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.107419
