670 likes | 2.28k Views
Binary Particle Swarm Optimization (PSO). Particle m. …. Particle 1. The Flowchart of Binary PSO. Generate and initialize particles with random position (X) and velocity (V). Evaluate position (Fitness). Update Position. If fitness(X) >fitness(Pbest) Pbest=X.
E N D
Particle m ….. Particle 1 The Flowchart of Binary PSO Generate and initialize particles with random position (X) and velocity (V) Evaluate position (Fitness) Update Position If fitness(X) >fitness(Pbest) Pbest=X If fitness(X) >fitness(Gbest) Gbest=X Update velocity Termination criterion is met? (e.g., Gbest=sufficient good fitness or maximum generations) Yes No Return the best solution
A Binary PSO A Particle Position vector, (m is the total number of particles). (n is the dimension of data). Velocity vector, is limited by
A Binary PSO • A particle = a solution or a gene subset. • If bit is 1,gene is selected. If bit is 0,gene is unselected. Particle position Gene expression data A subset of selected genes by a particle An example of a particle position representation in PSO for gene selection.
A Binary PSO Updating the velocity of a particle: Inertia W = inertial weight. = velocity for particle i at dimension d. Personal influence = acceleration constant. = random value. = position for particle i at dimension d. = the best previous position of the ith particle. Global influence = acceleration constant. = random value. = the global best position of all particles. Updating the position of a particle: if = random value. else
An Improved Binary PSO (IPSO) Idea Action Position update A new and simple rule Based on the whole of bits of a particle (Not based on single bit) Velocity update Particle velocity should be positive
1) if => or 2) if => or An Improved Binary PSO (IPSO) Analyzing the sigmoid function: The properties of the sigmoid function 3) if => or
An Improved Binary PSO (IPSO) 1) Modify the rule of position update: • The diagnostic goal = to develop a medical procedure based on the least number of possible genes for accurate disease detection. • Many previous works (biological and computational researches) have proved that a smaller number of genes can possible to produce higher classification accuracy. A new and simple rule of position update: if = random value. The whole of bits of a particle else
An Improved Binary PSO (IPSO) 2) A simple modification of the formula of velocity update The whole of bits of particles Calculation for the distance of two position. Example: and Step 1) Calculate the difference of bits for a = 4 b = 3 Step 2) Calculate the distance between and
An Improved Binary PSO (IPSO) 3) A Fitness Function: is leave-one-out-cross-validation (LOOCV) accuracy on the training set using the only genes in is the number of selected genes in M is the total number of genes for each sample and are two priority weights corresponding to the importance of accuracy and the number of selected genes, respectively.