Multi-label rules for phishing classification

摘要: Abstract Generating multi-label rules in associative classification (AC) from single label data sets is considered a challenging task making the number of existing algorithms for this rare. Current AC produce only largest frequency class connected with rule training set and discard all other classes even though these have representation rule’s body. In paper, we deal above problem by proposing an algorithm called Enhanced Multi-label Classifiers based Associative Classification (eMCAC). This discovers associated that current are unable to induce. Furthermore, eMCAC minimises extracted using classifier building method. The proposed has been tested on real world application related website phishing results reveal eMCAC’s accuracy highly competitive if contrasted known classic mining. Lastly, experimental show our able derive new end-users can exploit decision making.