An improved immune algorithm with parallel mutation and its application

The objective of this paper is to design a fast and efficient immune algorithm for solving various optimization problems. The immune algorithm (IA), which simulates the principle of the biological immune system, is one of the nature-inspired algorithms and its many advantages have been revealed. Although IA has shown its superiority over the traditional algorithms in many fields, it still suffers from the drawbacks of slow convergence and local minima trapping problems due to its inherent stochastic search property. Many efforts have been done to improve the search performance of immune algorithms, such as adaptive parameter setting and population diversity maintenance. In this paper, an improved immune algorithm (IIA) which utilizes a parallel mutation mechanism (PM) is proposed to solve the Lennard-Jones potential problem (LJPP). In IIA, three distinct mutation operators involving cauchy mutation (CM), gaussian mutation (GM) and lateral mutation (LM) are conditionally selected to be implemented. It is expected that IIA can effectively balance the exploration and exploitation of the search and thus speed up the convergence. To illustrate its validity, IIA is tested on a two-dimension function and some benchmark functions. Then IIA is applied to solve the LJPP to exhibit its applicability to the real-world problems. Experimental results demonstrate the effectiveness of IIA in terms of the convergence speed and the solution quality.

A survey of evolutionary algorithms for supervised ensemble learning

This paper presents a comprehensive review of evolutionary algorithms that learn an ensemble of predictive models for supervised machine learning (classification and regression). We propose a detailed four-level taxonomy of studies in this area. The first level of the taxonomy categorizes studies based on which stage of the ensemble learning process is addressed by the evolutionary algorithm: the generation of base models, model selection, or the integration of outputs. The next three levels of the taxonomy further categorize studies based on methods used to address each stage. In addition, we categorize studies according to the main types of objectives optimized by the evolutionary algorithm, the type of base learner used and the type of evolutionary algorithm used. We also discuss controversial topics, like the pros and cons of the selection stage of ensemble learning, and the need for using a diversity measure for the ensemble’s members in the fitness function. Finally, as conclusions, we summarize our findings about patterns in the frequency of use of different methods and suggest several new research directions for evolutionary ensemble learning.

Dynamic Uncertain Causality Graph for Knowledge Representation and Reasoning: Utilization of Statistical Data and Domain Knowledge in Complex Cases

The dynamic uncertain causality graph (DUCG) is a newly presented framework for uncertain causality representation and probabilistic reasoning. It has been successfully applied to online fault diagnoses of large, complex industrial systems, and decease diagnoses. This paper extends the DUCG to model more complex cases than what could be previously modeled, e.g., the case in which statistical data are in different groups with or without overlap, and some domain knowledge and actions (new variables with uncertain causalities) are introduced. In other words, this paper proposes to use -mode, -mode, and -mode of the DUCG to model such complex cases and then transform them into either the standard -mode or the standard -mode. In the former situation, if no directed cyclic graph is involved, the transformed result is simply a Bayesian network (BN), and existing inference methods for BNs can be applied. In the latter situation, an inference method based on the DUCG is proposed. Examples are provided to illustrate the methodology.

Dynamic Uncertain Causality Graph for Knowledge Representation and Probabilistic Reasoning: Directed Cyclic Graph and Joint Probability Distribution

Probabilistic graphical models (PGMs) such as Bayesian network (BN) have been widely applied in uncertain causality representation and probabilistic reasoning. Dynamic uncertain causality graph (DUCG) is a newly presented model of PGMs, which can be applied to fault diagnosis of large and complex industrial systems, disease diagnosis, and so on. The basic methodology of DUCG has been previously presented, in which only the directed acyclic graph (DAG) was addressed. However, the mathematical meaning of DUCG was not discussed. In this paper, the DUCG with directed cyclic graphs (DCGs) is addressed. In contrast, BN does not allow DCGs, as otherwise the conditional independence will not be satisfied. The inference algorithm for the DUCG with DCGs is presented, which not only extends the capabilities of DUCG from DAGs to DCGs but also enables users to decompose a large and complex DUCG into a set of small, simple sub-DUCGs, so that a large and complex knowledge base can be easily constructed, understood, and maintained. The basic mathematical definition of a complete DUCG with or without DCGs is proved to be a joint probability distribution (JPD) over a set of random variables. The incomplete DUCG as a part of a complete DUCG may represent a part of JPD. Examples are provided to illustrate the methodology.

SSC: a classifier combination method based on signal strength

We propose a new classifier combination method, the signal strength-based combining (SSC) approach, to combine the outputs of multiple classifiers to support the decision-making process in classification tasks. As ensemble learning methods have attracted growing attention from both academia and industry recently, it is critical to understand the fundamental issues of the combining rule. Motivated by the signal strength concept, our proposed SSC algorithm can effectively integrate the individual vote from different classifiers in an ensemble learning system. Comparative studies of our method with nine major existing combining rules, namely, geometric average rule, arithmetic average rule, median value rule, majority voting rule, Borda count, max and min rule, weighted average, and weighted majority voting rules, is presented. Furthermore, we also discuss the relationship of the proposed method with respect to margin-based classifiers, including the boosting method (AdaBoost.M1 and AdaBoost.M2) and support vector machines by margin analysis. Detailed analyses of margin distribution graphs are presented to discuss the characteristics of the proposed method. Simulation results for various real-world datasets illustrate the effectiveness of the proposed method.

Performance evaluation of multilayer perceptrons for discriminating and quantifying multiple kinds of odors with an electronic nose

This paper studies several types and arrangements of perceptron modules to discriminate and quantify multiple odors with an electronic nose. We evaluate the following types of multilayer perceptron. (A) A single multi-output (SMO) perceptron both for discrimination and for quantification. (B) An SMO perceptron for discrimination followed by multiple multi-output (MMO) perceptrons for quantification. (C) An SMO perceptron for discrimination followed by multiple single-output (MSO) perceptrons for quantification. (D) MSO perceptrons for discrimination followed by MSO perceptrons for quantification, called the MSO-MSO perceptron model, under the following conditions: (D1) using a simple one-against-all (OAA) decomposition method; (D2) adopting a simple OAA decomposition method and virtual balance step; and (D3) employing a local OAA decomposition method, virtual balance step and local generalization strategy all together. The experimental results for 12 kinds of volatile organic compounds at 85 concentration levels in the training set and 155 concentration levels in the test set show that the MSO-MSO perceptron model with the D3 learning procedure is the most effective of those tested for discrimination and quantification of many kinds of odors.