Sie sind auf Seite 1von 3

An Improved Genetic Algorithm and Its Blending Application with Neural Network

In order to overcome the limitation such as premature convergence and low global convergence speed of standard genetic algorithm, an improved genetic algorithm named adaptive genetic algorithm simulating human reproduction mode is proposed. The genetic operators of this algorithm include selection operator, help operator, adaptive crossover operator and adaptive mutation operator. The genetic individuals' sex feature, age feature and consanguinity feature are considered. Two individuals with opposite sex can reproduce the next generation if they are distant consanguinity individuals and their age is allowable. Based on this improved genetic algorithm, a thoroughly evolutionary neural network algorithm named IGA-BP algorithm is proposed. In IGA-BP algorithm, genetic algorithm is used firstly to evolve and design the structure, the initial weights and thresholds, the training ratio and momentum factor of neural network roundly. Then, training samples are used to search for the optimal solution by the evolutionary neural network. IGA-BP algorithm was used in pattern recognition example of Gray code. The illustrational results showed that IGA-BP algorithm was better than traditional neural network algorithm in both speed and precision of convergence, and its validity was proved.

A Hybrid Algorithm Based on Genetic Algorithm and Simulated Annealing for Solving Portfolio Problem
A hybrid algorithm is proposed based on genetic algorithm and simulated annealing for solving portfolio investment with probability criterion. Series structure and parallel structure are adopted by simulated annealing and genetic algorithm respectively. A hybrid algorithm is proposed for which generic algorithm is provided as the main frame of the parallel search, and simulated annealing is used in the mutation process of genetic algorithm. The hybrid algorithm holds the series-parallel structure, which enhance its ability to obtain the optimal solution in the whole solution space. The procedure of hybrid algorithm is described in detail. A case investigation shows that the hybrid algorithm is more feasible and effective by comparison with genetic algorithm. The proposed hybrid algorithm provides investors an applicable method to solve the portfolio problem.

A Genetic Algorithm-Based Approach to Flexible Job-Shop Scheduling Problem

Flexible Job-shop Scheduling Problem (FJSP) is one of extremely hard problems because it requires very large combinatorial search space. Genetic algorithm is wildly used to solve Flexible Job-shop Scheduling Problem. This paper presents an improved genetic algorithm. The improved genetic algorithm we proposed uses many different strategies to get a better result. During the phase of create initial population, the improved genetic algorithmtakes into account the number of operations in each job. And the intelligent mutation strategy is used which makes every individual and gene have different probability to mutate. In this paper, the object of schedulingalgorithm is to get a sequence of the operations on machines to minimize the makespan. And the performance of the improved genetic algorithm is compared with another genetic algorithm. During the experiment, the two improvements are compared respectively with the compared genetic algorithm. The results show that the improved genetic algorithm outperforms the compared algorithm.

A Path Adjustment Algorithm of the Combination of Genetic Algorithm and Ant Colony Algorithm
Under the premise of using Ford algorithm to obtain the shortest path in the link map, in this paper the path adjustment algorithm which is the combination of the genetic algorithm and ant colony algorithm is introduced. And the path adjustment algorithm is used to adjust and optimize the current shortest path. The definition of the genetic algorithm control function was to control the appropriate combination opportunity of the two algorithms. Taking advantage of genetic algorithm to produce the original results, they were transformed into the initial pheromones value needed by ant colony algorithm, then ant colony algorithm to get the best results. Through numerical simulation analysis, it is further description that the path adjustment algorithm's effectiveness and necessity.

Application of PSO-RBF Neural Network in Network Intrusion Detection


Detecting all kinds of intrusions efficiently is significant to network security. Radial basis function (RBF) neural network is a kind of feed forward neural network, which is widely employed as a real-time pattern classification. In RBF neural network, the center of radial basis function, the variance of radial basis of function and the weight have to be chosen. If they are not appropriately chosen, the RBF neural network may degrade validity and accuracy of modeling. Particle swarm optimization algorithm (PSO) is a member of the wide category of swarm intelligence methods to solve non-linear programming problems. PSO has proved to be competitive with genetic algorithm (GA) in parameter optimization. So PSO is used to optimize the RBF neural network parameters in this work. Therefore, the novel combination method based on RBF neural network and PSO (PSO-RBFNN) is adapted to network intrusion detection. The experimental results show that the proposed model is superior to the conventional RBF neural network.

Training back propagation neural networks with genetic algorithm for weather forecasting

Accurate weather forecasting is important in today's world as agricultural and industrial sectors are largely dependent on the weather conditions. Secondly, it is used to warn about natural disasters. Due to non-linearity in climatic physics, neural networks are suitable to predict these meteorological processes. Back Propagationalgorithm using gradient descent method is the most important algorithm to train a neural network for weather forecasting. Back propagation algorithm suffers from several problems. In this paper, in order to overcome some of these problems, an integrated back propagation based genetic algorithm technique to train artificial neuralnetworks is proposed. In the proposed technique, back propagation is combined with genetic algorithm in such a way that the pitfalls of the algorithm get converted to benefits. The results and comparison of the technique with the previous one are enlisted along with.

Swarm Intelligence in Vehicle Routing Swarm Intelligence in Optimal Resource Allocation Swarm Intelligence based Predictions

A Quantification Method for Network Security Situational Awareness Based on Conditional Random Fields
Network security situational awareness(NSSA) has been a hot research spot in the network security domain. In this paper, a quantification method for NSSA based on conditional random fields(CRFs) was proposed. The data of network attacks from intrusion detection system (IDS), the hosts' vulnerabilities and the hosts' states were firstly combined as the network security factors. And then the network security threat degree was defined to quantify the risk of the whole network and classify the attacks. A diverse set of effective features were incorporated in CRFs Model. Finally the experiments on the DARPA 2000 data set generate the explicit network security situational graph. It proves that the method introduced in this paper can represent network risk more accurate and offer a good quantification for the network security situation.

Identifying Brain Signals based on Neural Networks

Das könnte Ihnen auch gefallen