Classification Based on Pruning and Double Covered Rule Sets for the Internet of Things Applications

Algorithm 3

Inducing rule set 𝐴.

Input: Training data 𝑇_𝑖, CV_𝑖

Output: Rule set 𝐴

Method:

(1)  Rule set 𝑅=𝜙, dataset=𝜙, cover value cv=𝜙, IniteQueue 𝑄;

(2)  while CV_𝑖=𝜙

(3)   𝑄.push(𝑎) where 𝑎 is first element of CV_𝑖;

(4)   CV_𝑖=CV_𝑖−{𝑎};

(5)  end while

(6)  while !𝑄.𝑒𝑚𝑝𝑡𝑦()

(7)   pattern 𝑥=𝑄.front(); 𝑄.pop();

(8)   if 𝑥.length>max_rule_length continue;

(9)   compute the information gain of 𝑥;

(10)  if 𝑥.informationgain==0

(11)  𝑅=𝑅∪{𝑥};

(12)  else

(13)  dataset=dataset∪each𝑡.contains(𝑥);

(14)  found cv in dataset;

(15)  connect 𝑥 with cv;

(16)  𝑄.push();

(17)  dataset=𝜙, cv=𝜙;

(18)  end if

(19)  end while

(20)  return 𝑅