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

Algorithm 4

Inducing rule set 𝐵.

Input: Training data 𝑇_𝑖, CV_𝑖

Output: Rule set 𝐵

Method:

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

(2)  add cover value which can cover 𝑇_𝑖 to cs, and cs∉CV_𝑖

(3)  while cs≠𝜙

(4)       𝑄.push(𝑎) where 𝑎 is first element of cs;

(5)       cs=cs−{𝑎};

(6)  end while

(7)  while !𝑄.empty()

(8)   𝑥=𝑄.front(); 𝑄.pop()

(9)   If 𝑥.length>max_rule_length continue;

(10)     compute the information gain of 𝑥;

(11)     if 𝑥.informationgain==0

(12)     𝑅=𝑅∪{𝑥};

(13)     else

(14)     dataset=dataset∪each𝑡.Contains(𝑥);

(15)     Found cv in dataset, and cv∉CV_𝑖;

(16)     connect 𝑥 with cv;

(17)     𝑄.push();

(18)    dataset=𝜙, cv=𝜙;

(19)     end if

(20)  end while

(21)  return 𝑅