Research Article
The Reputation Evaluation Based on Optimized Hidden Markov Model in E-Commerce
Algorithm 1
Hidden Markov Model Algorithm based on Particle Swarm Optimization.
Input: Observation sequence, the number of hidden states | Initialization: initialized particles and corresponding velocities | randomly | while a termination criterion is false do | Compute the fitness values for all particles in swarm with (18) | Find the local optimal solution () | Find the global optimal solution () | Update all velocities with (16) | Update all particles with (17) | Re-mapping particles with (20) | Adjust the velocities of the re-mapped particles with (21) | Re-normalization particles with (22) | end | Output: optimized model parameters . |
|