A Novel Approach for Classifying MANETs Attacks with a Neutrosophic Intelligent System based on Genetic Algorithm

Algorithm 2

SOFM algorithm for generating the membership, nonmembership, and indeterminacy functions of the neutrosophic variable.

Input: input_data vectors(Training_data set), Input_dim, output_dim,

Output: neutrosophic variable membership, nonmembership and indeterminacy functions

//membership function generation

   Training_DataRead_data(membership_data)

   Membership_data SOFM(Trainig_data, Input_dim, output_dim)

   Draw (Membership_data)

   Training_DataRead_data(nonmembership_data)

   Non_Membership_data SOFM(Trainig_data, Input_dim, output_dim)

   Draw (Non_Membership_data)

   Indeterminancy Calculate_ind(Membership_data, Non_Membership_data)

   Draw (Indeterminancy)

End

Function SOFMFunction Update_weights

Input: Trainig_data, Input_dim, output_dim Input: Winning_neuronq_j

Output: Output _Function Output: Update_weights

  Initialize_SOFM (input_neurons, output_neurons) Find (Winning_neuronq_j)

  Randomly_Initialize_SOFM_Weights ()   

  While Error>threshold Do

  Foreach Record in Training_Data (3)

  Input_Record ();

  Winning_neuronq_j=; (4) Output (Update_weights)

  Update_weights (Winning_neuronq_j); (5) End fun

  Endforeach

  Error =Calculate_ErrorRate ();

  End while

  Retrieving_phase ();

  Output FunctionNetwork_Weights)

End fun