meta-heuristic algorithms

Today, due to the considerable benefits of the Internet of Things (IoT) in various fields such as smart homes, industry, cars, agriculture, etc., its application is very widespread. Due to this, the security of these networks is receiving more and more attention. One of the methods of providing security in networks as well as IoT network is intrusion detection systems. Traditional intrusion detection systems are not very efficient for use in the Internet of Things, so the use of new methods is required. One of these methods is intrusion detection systems based on machine learning and deep learning that have been considered in this area. They are trained in machine learning and deep neural network learning to detect attack patterns. There are important parameters for setting up a machine learning network, and choosing the right value for these parameters has a great impact on system accuracy. In this paper, a method is presented that uses meta-heuristic algorithms such as genetic algorithm, particle swarm optimization, artificial bee colony and gray wolf to find the optimal hyperparameters for the deep learning network and the intrusion detection system is created based on these hyperparameters. This method was implemented using the Tensorflow and keras libraries and tested on the KDDCup99, UNSW-NB15 and Bot-IoT datasets. The results showed that the proposed method can detect attacks with a high accuracy of 99%.

The use of energy-conscious solutions is one of the important research topics in the field of cloud computing. By effectively using virtual machine placement and aggregation algorithms, cloud suppliers will be able to reduce energy consumption. In this paper, a new model is presented that seeks to achieve the desired results by improving the algorithms and providing appropriate methods. Periodic monitoring of resource status, proper analysis of the data obtained, and prediction of the critical state of the servers using the proposed Markov model have reduced the number of unnecessary migrations as much as possible. The combination of genetic algorithm and simulated annealing in the replacement section along with the definition of the adsorbent Markov chain has resulted in better and faster performance of the proposed algorithm. Simulations performed in different scenarios in CloudSim show that compared to the best algorithm compared, at low, medium and high load, energy consumption has decreased significantly. Violations of service level agreements also fell by an average of 17 percent.

In the past decades, vehicle routing problem (VRP) has gained considerable attention for its applications in industry, military, and transportation applications. Vehicle routing problem with simultaneous pickup and delivery is an extension of the VRP. This problem is an NP-hard problem; hence finding the best solution for this problem which is using exact method, take inappropriate time, and these methods are not useful in real-world applications. Using meta-heuristic algorithms for calculating and computing the solutions for NP-hard problems is a common method to contrast this challenge. The objective function defined for this problem, is a constrained objective function. In previous algorithms, the penalty method was used as constraint handling technique to define the objective function. Determining the value of parameters and penalty coefficient is not easy in these methods. Moreover, the optimal number of vehicles was not considered in the previous algorithms. So, the user should guess number of vehicles and compare the result with other values for this variable. In this paper, a novel objective function is defined to solve the vehicle routing problem with simultaneous pickup and delivery. This method can find the vehicle routes such that increases the performance of the vehicles and decreases the processes’ costs of transportation. in addition, the optimal number of vehicle in this problem can be calculated using this objective function. Finding the best solution for this optimization problems is an NP-hard and meta-heuristic methods can be used to estimate good solutions for this problem. Then, a constrained version of gravitational search algorithm is proposed. In this method, a fuzzy logic controller is used to calculate the value of the parameters and control the abilities of the algorithm, automatically. Using this controller can balance the exploration and exploitation abilities in the gravitational search algorithm and improve the performance of the algorithm. This new version of gravitational search algorithm is used to find a good solution for the predefined objective function. The proposed method is evaluated on some standard benchmark test functions and problems. The experimental results show that the proposed method outperforms the state-of-the-art methods, despite the simplicity of implementation.

Parameters of meta-heuristic algorithms are very effective in their performance and are usually done experimentally, which is very time-consuming. In this research, a hybrid method for selecting the optimal parameters of meta-heuristic algorithms is presented. The proposed method is a combination of data envelopment analysis methods and response surface methodology and is called DSM. In fact, this method can be used to optimize multi-objective problems and its main advantage is to create and optimize one performance response procedure instead of optimizing multiple output response procedures. In addition to optimizing parameters, it also simultaneously maximizes efficiency. In this research, the proposed DSM method has been used to adjust the parameters of the cuckoo optimization algorithm to optimize the standard and experimental Aklay and Rastrigin functions. In the hybrid DSM method, first, the efficiency value is calculated using data envelopment analysis for each set of meta-heuristic algorithm parameters, then the response procedure for performance is determined according to the meta-heuristic algorithm parameters using the response surface methodology. Finally, by optimizing the efficiency surface, the optimal values of the cuckoo algorithm parameters are obtained. In order to validate, the results of the proposed method have been compared with a similar method. The results show better performance of the hybrid algorithm in terms of solution time, number of iterations, and accuracy of the optimization function compared to other similar methods.

One of the major challenges in cyber space is the existence of fake or phishing pages that needs the attention of command control systems. In phishing attacks, people are directed to fake pages and their important information is stolen by a thief or phisher. Machine learning and data mining algorithms are the widely used algorithms for phishing websites classification. Feature selection has a great influence on the classification results. In this research, an improved spotted Hyena optimization algorithm (ISHOA) is proposed to select appropriate features for classifying phishing websites through artificial neural network. The proposed ISHOA outperformed the standard spotted Hyena optimization algorithm with 98.64% better accuracy. In addition, the results indicate the superiority of ISHOA to three other meta-heuristic algorithms including: particle swarm optimization, firefly algorithm, and bat algorithm. The proposed algorithm is also compared with a number of classification algorithms proposed before on the same dataset and its dominance is showed.

In multi-label data, each instance is associated with a set of labels, instead of one label. Due to the increasing number of modern applications associated with multi-label data, multi-label classification has gained significant attention during recent years. As in single-label data, eliminating redundant and/or irrelevant features plays an important role in improving classification performance. In this paper, a hybrid method for multi-label feature selection problem based on combing filter and wrapper methods is proposed, where meta-heuristic algorithms are employed as the wrapper method. Since, the number of features in multi-label data is usually high, solely employing search algorithms for finding the optimal feature subsets has high computational burden, and is very possible to fail. Hence, irrelevant features are first detected and removed by a filter method. Then, salient features are found among the remained features by the help of meta-heuristic algorithms. A significant number of well-known meta-heuristic algorithms are employed as the wrapper method, in the proposed system. Experiments show that the proposed method obtains better classification results, compared to other algorithms.

In recent years, research works in field of the network security have been directed to inspire from biological immune system so as to solve complex problems. Artificial immune system (AIS) and its applied immunity potential with prerequisite for bio defense, is always involved as a method for organization’ security control and network anomaly detection. In this research, different immunity methods in comparison with other machine learning and meta-heuristic algorithms have been analyzed for our main purpose; that is to provide a novel approach for solving the intrusion detection. All of evaluations accomplished on WEKA data mining tool v3.6 under NSL-KDD dataset. Results of experiment show that the AIS methods ARIS2Parallel, Immunos99 and CSCA cause to enhance in accuracy rates sensibly after feature selection phase was embedded to them. So, approach of Bat+ARIS2Parallel with correlation coefficient of 0.946, detection rate of 0.973, accuracy rate of 0.9725 and false positive rate of 0.028 has obtained an idealistic classification compared to other approaches in tests. Since it has high cc rate, it seems that is reliable to be used in researches for IDS developments in future.