A Modified Decision Tree Algorithm Based on Genetic Algorithm for Mobile User Classification Problem

Algorithm 2

Algorithm description.

Algorithm: Decision tree_Genetic algorithm

Input:

  Data set R, parameters for Genetic algorithm

Output:

  Optimal classification rules

Begin:

  𝐼=0;

 Initialize P(I);

   // initialize population; Scan 𝑅, select records where class attribute value = 𝐶_𝑖

Preprocessing 𝑅;

   //this process including clean data, discretize continuous attributes, calculate each feature attribute information gain rate,

  code recorded data and get initial population.

Fitness P(I);

Avg(Fitness P(I));

While (I <= Max_generation or Avg_𝑖 − Avg_𝑖−1 > 𝜀)

{

     I ++;

     GA-Operation P(I);

     Fitness P(I);

   }

Or delete the individual which fitness less than the threshold value;

Optimal Population P(I);

 //get the optimal classification rules and calculate the average the Fitness(I)

END