Ant colony optimization algorithms have been applied to many combinatorial optimization problems, ranging from quadratic assignment to protein folding or routing vehicles and a lot of derived methods have been adapted to dynamic problems in real variables, stochastic problems, multi-targets and parallel implementations. It has also been used to classification of the data set based on the attribute. It has been observed that construct solution and pheromone update play an important role in the ACO algorithm. The selection of the pheromone update is based on the construct solution which is further base on the probability function and initial selection. So if the selection of the pheromone done properly then ACO algorithm will terminate in less number of the iteration and it will be produce the good result. It has further observed that difference result have been possible for the different selection of the construct and pheromone on the same data set. Therefore, in this paper an effort has been made to suggest the techniques to select the initial construct and pheromone update for data set and the classification has to be done using the concept of clustering.

An fast ant colony algorithm solving traveling salesman problem is presented. Firstly, this paper gives a new pheromone updating model. It reduces the complexity of the search process, improves the accuracy of the route searching. Secondly, by setting penalty function, to exclude irrelevant path so that the narrow scope of the search process. Experiments show that, this algorithm can get better optimal solution, and improve the convergence speed significantly.

Ant colony optimization (ACO) is a population-based metaheuristic technique to solve combination optimization problems effectively. However, how to improve the performance of ACO algorithms is still an active research topic. Though there are many algorithms solving TSPs effectively, there is an application bottleneck that the ACO algorithm costs too much time in order to get an optimal solution. This paper revised pheromones in local and global update mode--a fast ACO algorithm for solving TSPs is presented in this paper. Firstly, a new pheromone increment model called ant constant, which keeps energy conversation of ants, is introduced to embody the pheromone difference of different candidate paths. Meanwhile, a pheromone diffusion model, which is based on info fountain of a path, is established to reflect the strength field of the pheromone diffusion faithfully, and it strengthens the collaboration among ants. Experimental results on different benchmark data sets show that the proposed algorithm can not only get better optimal solutions but also enhance greatly the convergence speed.

PSO algorithm is easy to operate and to realize, and it has got many scholars' attention once proposed. In recent years there has appeared many improved PSO algorithm, but it cann't get the global optimal answer in probability 1. Thinking over the problem using probability theory, this paper can achieve the optimal answer as far as possible big Priori probability, and experiments show that the improved PSO algorithm has avoided local minima successfully and got higher search rates.

A particle swarm optimization algorithm based on the solution space division was presented.On the premise of no impact on the particle swarm optimization,this algorithm can preprocess the solution space in order to get the optimal-intervals and improve the particle swarm optimization algorithm's efficiency.Meanwhile,the checkpoints would be set up to update the solution space division dynamically.Experiments show that the algorithm can effectively improve the particle swarm optimization algorithm's efficiency and solve the problem of trapping the local minimum.Furthermore,the algorithm can get minimum iteration times by designated precision and the results are satisfactory in the adaptive state.

A Particle Swarm Optimization Algorithm Based on the Solution Space Division is presented. No impact on the Particle Swarm Optimization, this algorithm can preprocess the solution space in order to get the optimal-intervals and improve the Particle Swarm Optimization Algorithm's efficiency. Meanwhile, the checkpoints would be set up to update the Solution Space Division dynamically. Experiments show that the algorithm can effectively improve the Particle Swarm Optimization Algorithm's efficiency and the problem of trapping the local minimum. Furthermore, the algorithm can get minimum iteration times by designated precision and the results are satisfactory in the adaptive state.