Constrained dynamic rule induction learning

作者：

Highlights：

• New dynamic rule induction method is presented.

• Major modifications to PRISM method are proposed.

• Applicability of rule induction in particular eDRI and PRISM on real phishing data sets.

• Experimental results analysis using different algorithms was performed.

• Two sets of data are used to measure the pros and cons (UCI and security).

摘要

•New dynamic rule induction method is presented.•Major modifications to PRISM method are proposed.•Applicability of rule induction in particular eDRI and PRISM on real phishing data sets.•Experimental results analysis using different algorithms was performed.•Two sets of data are used to measure the pros and cons (UCI and security).

论文关键词：Classification,Data mfining,Prediction,PRISM,Rule induction,Online security

论文评审过程：Received 24 February 2015, Revised 29 March 2016, Accepted 19 June 2016, Available online 24 June 2016, Version of Record 1 July 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.06.041