Fuzziness in Ant Colony Optimization and Their Applications

Nature-inspired algorithms are still at a very early stage with a relatively short history, comparing with many traditional, well-established methods. Metaheuristics, in their original definition, are solution methods that orchestrate an interaction between local improvement procedures and higher level strategies to create a process capable of escaping from local optima and performing a robust search of a solution space. One major algorithm is Ant Colony Optimization which has been applied in varied domains to better the performance. Fuzzy Linear Programming models and methods has been one of the most and well-studied topics inside the broad area of Soft Computing. Its applications as well as practical realizations can be found in all the real-world areas. Here we wish to introduce how fuzziness can be included in a nature inspired algorithm like ant colony optimization and thereby enhance its functionality. Several applications of ACO with fuzzy concepts will be introduced in the chapter.

Nature-Inspired-Algorithms-Based Cellular Location Management

Nature-Inspired algorithms have gained relevance particularly for solving complex optimization problems in engineering domain. An overview of implementation modeling of the established algorithms to newly developed algorithms is outlined. Mobile location management has vital importance in wireless cellular communication and can be viewed as an optimization problem. It has two aspects: location update and paging where the objective is to reduce the overall cost incurred corresponding to these two operations. The potential application of the Nature-Inspired algorithms to mobile location management is studied. Many such algorithms are recently being explored along with incremental modifications to the existing techniques. Finally, analysis and insights highlight the further scopes of the Nature-Inspired algorithms to mobile location management application.

DE-Based RBFNs for Classification With Special Attention to Noise Removal and Irrelevant Features

This chapter presents a novel approach for classification of dataset by suitably tuning the parameters of radial basis function networks with an additional cost of feature selection. Inputting optimal and relevant set of features to a radial basis function may greatly enhance the network efficiency (in terms of accuracy) at the same time compact its size. In this chapter, the authors use information gain theory (a kind of filter approach) for reducing the features and differential evolution for tuning center and spread of radial basis functions. Different feature selection methods, handling missing values and removal of inconsistency to improve the classification accuracy of the proposed model are emphasized. The proposed approach is validated with a few benchmarking highly skewed and balanced dataset retrieved from University of California, Irvine (UCI) repository. The experimental study is encouraging to pursue further extensive research in highly skewed data.

A Holistic-Based Multi-Criterion Decision-Making Approach for Solving Engineering Sciences Problem Under Imprecise Environment

Multi-Criteria Decision Making has evolved as an important tool for taking some of the most important decisions in the today's hi-tech engineering world. But due to some reasons like measurement difficulty, lack of data, faulty instruments, etc., or due to lack of absolute information about the topic, alternatives present and the criteria, decision making becomes very difficult as all parameter for modeling a decision making problem are not precise. In such scenario the importance of one with respect to the others are represented in terms of linguistic factor. Such cases could be tackled by considering the problem in fuzzy environment. In this chapter, the different hybrid fuzzy MCDM techniques are shown along with their application in different engineering problems. One problem is randomly selected and solved using different fuzzy MCDM techniques and compared the result with the existing literature.

Applications of Hybrid Intelligent Systems in Adaptive Communication

Dynamic allocation of the resources for optimum utilization and throughput maximization is one of the most important fields of research nowadays. In this process the available resources are allocated in such a way that they are maximally utilized to enhance the overall system throughput. In this chapter a similar problem is approached which is found in Orthogonal Frequency Division Multiplexing (OFDM) environment, in which the transmission parameters namely the code rate, modulation scheme and power are adapted in such a way that overall system's data rate is maximized with a constrained bit error rate and transmit power. A Fuzzy Rule Base System (FRBS) is proposed for adapting the code rate and modulation scheme while Genetic Algorithm (GA) and Differential Evolution (DE) algorithm are used for adaptive power allocation. The proposed scheme is compared with other schemes in the literature including the famous Water-filling technique which is considered as a benchmark in the adaptive loading paradigm. Supremacy of the proposed technique is shown through computer simulations.

Development of an Efficient Prediction Model Based on a Nature-Inspired Technique for New Products

A lot of studies have been made on new product development process to make it an ideal procedure and many researchers have contributed significantly to achieve this by studying various factors associated with it. In this study, an attempt has been made to predict the optimal numbers of new products produced by electronics and metal & machinery industry by considering various factors those significantly affects the production pattern of these industries. For prediction purposes, functional linked artificial neural network (FLANN) with and without nature-inspired techniques have been used and comparison of performance for both the models have been done by using mean square error (MSE) and mean absolute percentage error (MAPE) as the measurement indices.

A Statistical Scrutiny of Three Prominent Machine-Learning Techniques to Forecast Machining Performance Parameters of Inconel 690

Computational approaches like “Black box” predictive modeling approaches are extensively used technique applied in machine learning operations of today. Considering the latest trends, present study compares capabilities of two different “Black box” predictive model like ANFIS and ANN with a population-based evolutionary algorithm GEP for forecasting machining parameters of Inconel 690 material, machined in a CNC-assisted 3-axis milling machine. The aims of this article are to represent considerable data showing, every techniques performance under the criteria of root mean square error (RSME), Correlational coefficient R and Mean absolute percentage error (MAPE). In this chapter, we vigorously demonstrate that the performance of the GEP model is far superior to ANFIS and ANN model.

A Novel Hybrid Model Using RBF and PSO for Net Asset Value Prediction

Net asset value (NAV) prediction is an important area of research as small investors are doing investment in there, Literature survey reveals that very little work has been done in this field. The reported literature mainly used various neural network models for NAV prediction. But the derivative based learning algorithms of these reported models have the problem of trapping into the local solution. Hence in chapter derivative free algorithm, particle swarm optimization is used to update the parameters of radial basis function neural network for prediction of NAV. The positions of particles represent the centers, spreads and weights of the RBF model and the minimum MSE is used as the cost function. The convergence characteristics are obtained to show the performance of the model during training phase. The MAPE and RMSE value are calculated during testing phase to show the performance of the proposed RBF-PSO model. These performance measure exhibits that the proposed model is better model in comparison to MLANN, FLANN and RBFNN models.

Swarm-Based Nature-Inspired Metaheuristics for Neural Network Optimization

Nature-inspired algorithms have been productively applied to train neural network architectures. There exist other mechanisms like gradient descent, second order methods, Levenberg-Marquardt methods etc. to optimize the parameters of neural networks. Compared to gradient-based methods, nature-inspired algorithms are found to be less sensitive towards the initial weights set and also it is less likely to become trapped in local optima. Despite these benefits, some nature-inspired algorithms also suffer from stagnation when applied to neural networks. The other challenge when applying nature inspired techniques for neural networks would be in handling large dimensional and correlated weight space. Hence, there arises a need for scalable nature inspired algorithms for high dimensional neural network optimization. In this chapter, the characteristics of nature inspired techniques towards optimizing neural network architectures along with its applicability, advantages and limitations/challenges are studied.

Metaheuristic-Based Hybrid Feature Selection Models

This chapter focuses on key applications of metaheuristic techniques in the field of gene selection and classification of microarray data. The metaheuristic techniques are efficient in handling combinatorial optimization problems. In this article, two different types of metaheuristics such as Genetic algorithm (GA) and Particle Swarm Optimization (PSO) are hybridized with fuzzy-rough (FR) method for optimizing the subset selection process of microarray data. The FR method applied here deals with impreciseness and uncertainty of microarray data. The predictive accuracy of the models is evaluated by an adaptive neural net ensemble and by a rule based classifier MODLEM respectively. Moreover, the learning efficiency of the ensemble is compared with base learners and with two classical ensembles. The rule based classifier generates a set of rules for disease diagnosis and prognosis and enables to study the function of genes from gene ontology website. The experimental results of both the models prove that, hybrid metaheuristic techniques are highly effective for finding potential genes.