Abstract The paper illustrates the application of ant colony optimization algorithm to solve both continuous function and combinatorial problems in reliability engineering. is combined with strength Pareto fitness assignment procedure handle multiobjective problems. Further, a clustering has been applied prune set maintain diversity. Benchmark case examples show superiority such Copyright © 2002 John Wiley & Sons, Ltd.

This study presents an allergenic protein prediction system that appears to be capable of producing high sensitivity and specificity.The proposed is based on support vector machine (SVM) using evolutionary information in the form amino acid position specific scoring matrix (PSSM).The performance this assessed by a 10-fold crossvalidation experiment dataset consisting 693 allergens 1041 non-allergens obtained from Swiss-Prot Structural Database Allergenic Proteins (SDAP).The PSSM method...

In this work we propose a Pareto-based multi-objective search strategy for subgraph mining in structural databases. The method is an extension of Subdue, classical graph-based knowledge discovery algorithm, and it thus called MultiObjective Subdue (MOSubdue). MOSubdue incorporates the NSGA-II's crowding selection mechanism order to retrieve well distributed Pareto optimal set meaningful subgraphs showing different trade-offs between support complexity, single run. good performance proposed...

Abstract This article focuses on the development of a multiobjective optimization algorithm for tubular reactor–regenerator system with moving deactivating catalyst. The task is to find optimal temperature profile along reactor, catalyst recycle ratio, and regeneration capacity maximizing process profit flux, selectivity, conversion. A new heuristic technique, viz, ant colony method has been employed obtain Pareto set solutions. Keywords: Ant algorithmMultiobjective...

Subgraph Mining aims to find frequent, descriptive and interesting subgraphs in a graph database. Usually, this search involves simple user-defined thresholds is only driven by single-objective. In paper, we propose an Evolutionary Multiobjective Optimization algorithm, called MOEP-SO, mine from graph-represented data maximizing two objectives, support size of the subgraphs. Experimental results on synthetic real-life graph-based datasets validate utility proposed methodology when...

This work proposes multiobjective subgraph mining in graph-based data using evolutionary programming (MOEP). A mined is defined by two objectives, support and size. These objectives are conflicting as a with high value usually of small size vice-versa. MOEP applies NSGA-II's nondominated sorting procedure to evolve the population during generation process. An experimental study on five synthetic real-life datasets shows that outperforms Subdue-based methods, well-known heuristic search...

Scientograms are graph representations of scientific information. Exploring vast amount scientograms for data analysis has been great interest in Information Science. This work emphasizes the application multiobjective subgraph mining scientogram task regarding extraction common research areas world. For this task, we apply a recently proposed Subdue (MOSubdue) algorithm frequent graph-based data. The incorporates several ideas from evolutionary optimization. underlying structure is social...