Search

The number of scientific publications in different knowledge fields has considerably grown in recent times. This makes difficult for researchers to synthesize all the scientific-technical advances, so automatic summarization methods of scientific papers would be helpful. These methods generate a summary from a reference paper with its most relevant contributions. More specifically, citation-based summarization considers the citation contexts to the reference paper in subsequent publications. For the first time, this problem has been formulated as a multi-objective optimization problem, optimizing the content coverage and the redundancy reduction in a simultaneous way. A Decomposition-based Multi-Objective optimization algorithm for Citation-based Summarization (DMOCS) has been designed, developed, and applied for solving this problem. The results obtained by the proposed approach have improved the existing ones in the scientific literature between 17.47% and 133.50%, increasing the ROUGE percentage improvements when the N -gram is larger. Besides, an exhaustive analysis of the different parts of a scientific paper has been performed, showing that the citations from the citing papers with their corresponding spans in the reference paper impact in the quality of a citation-based summary. • First time the citation-based summarization is solved by multi-objective optimization. • A decomposition-based multi-objective optimization algorithm (DMOCS) is proposed. • An extensive analysis of the best text parts for citation-based summarization is done. • Interesting conclusions about the citing spans and the citation contexts are obtained. • DMOCS improves the results of the different approaches in the scientific literature. [ABSTRACT FROM AUTHOR]

Solving multi-objective optimization problems plays an important role in several applications. Recently, the Marine Predator Algorithm (MPA) was introduced for solving single objective optimization problems inspired by the behaviour of marine predator in their search for a prey. This paper proposes a modified MPA based framework called Guided Multi-objective Marine Predator Algorithm (GMOMPA) to solve multi-objective optimization problems. The proposed GMOMPA incorporates an external archive to help store the optimal Pareto set solution and guide the particles during the exploitation of the search space. For obtaining non-dominated solutions, the Pareto dominance concept is utilized while the epsilon dominance is considered to update the archive's sorted solutions. In this context, the epsilon dominance concept extends the diversity and exploration of the solutions. Further, a fast non-dominated solution and crowding distance are introduced to update the particle's position, while maintaining the diversity and ensuring a fast convergence towards the Pareto optimal. The proposed GMOMPA is evaluated on several different benchmark test functions including multi-objective ZDT, DTLZ, UF, and WFG test functions as well as the recent multi-objective multimodal CEC 2020 test functions. Moreover, the performance of the proposed GMOMPA is compared with well-known multi-objective optimization algorithms. The obtained results show that the proposed GMOMPA is a good tool for multi-objective optimization and has significant advantages over several state-of-the-art algorithms in almost all of the test functions. [ABSTRACT FROM AUTHOR]

For years, meta-heuristic algorithms have been widely studied and many improved versions have been developed: from the evolution of the swarm topologies of the Particle Swarm Optimization algorithm, to the using of machine learning to Differential Evolutionary algorithms. However, the tuning of the fundamental meta-heuristic parameters has been less studied, but may lead to significant improvements on the convergence accuracy of these algorithms. This paper aims at developing an automated methodology to calibrate the parameters of population-based meta-heuristic algorithms for optimization problems. Based on the kriging estimation of the best combination of parameters, the Automated parameter tuning of Meta-heuristics (AptM) methodology gives the optimal algorithm setup for each considered problem in order to lead to a better convergence accuracy. The proposed AptM methodology is used to tune three different meta-heuristic algorithms, each applied to twelve mathematical unimodal or multimodal objective functions. AptM methodology performance is assessed by comparison of classical setups usually used in the literature. The numerical results show that the AptM methodology allows a significant improvement of the convergence accuracy of meta-heuristics with an average improvement of 62.02%, 69.12% and 64.94% on optimization problems defined in dimensions 10, 30 and 50 respectively. An experimental criterion is defined based on the convergence accuracy of the AptM methodology over the classical setups, assessing the AptM performances. The previous experimental criterion allows to compare the AptM methodology over the base-set. The AptM methodology shows a significant improvement of the algorithms performance on 97.2% of the tested problems. [ABSTRACT FROM AUTHOR]

In recent years, brain storm optimization (BSO) algorithm has received much attention in solving classical optimization problems and is used to implement evolutionary classification models. However, in practical applications, large-scale datasets complicate the structure of the classification model, which can have a great impact on the classification performance. In the optimization process, the traditional single-strategy BSO cannot preserve the information of dominant solution well, and its generation strategy is inefficient in solving various complex practical problems. To solve this problem, we introduce feature selection to improve the optimization model structure. Meanwhile, in order to enhance the search capability of BSO, three new generation strategy are embedded in the BSO algorithm in this paper. With the three generation methods of global optimal, local optimal and nearest neighbor, the information of the dominant solution can be better preserved and the search efficiency can be improved. The performance of the proposed generation strategy in solving classification problems is demonstrated on ten datasets with different sizes and dimensions. The experimental results reveal that the new generation strategy can enhance the performance of BSO algorithm for solving classification problems. [ABSTRACT FROM AUTHOR]

This paper proposes a new multi-objective evolutionary algorithm by adapting the recent Henry Gas Solubility Optimization (HGSO) with multiple objectives. The proposed MOHGSO uses the Pareto dominance relation as means of comparison and integrates two types of archive, while an elite archive is used to store the Pareto solutions found over the evolutionary process, the other external-archives are used to store the local best solutions corresponding to each cluster. Moreover, efficient archiving and leader selection strategies based on the crowding distance computation are proposed to guide the population towards the true Pareto front. The performance of the MOHGSO algorithm is validated through an extensive comparison with three well-known algorithms on twelve test functions and four engineering design problems. The experiments of two widely used metrics in the field called IGD and Sp metrics show the ability of the proposed algorithm in achieving interesting results. Furthermore, the statistical results related to the Wilcoxon test indicate that the proposed algorithm outperforms significantly the selected methods for the above metrics at a 99% confidence level. [ABSTRACT FROM AUTHOR]

Coverage plays a vital role in the performance and proper functioning of wireless sensor networks. However, ensuring a network's coverage is met numerous challenges due to sensors having limited sensing range, communication range, and energy. Many coverage problems are NP-hard, one of which is the network coverage with lifetime problem (CTLP). As such, a number of meta-heuristic algorithms have been proposed to solve CTLP in practical scenarios. This paper proposes an approach for CTLP based on the teaching–learning based optimization algorithm (TLBO), which is often employed to address continuous optimization problems. Specifically, a discrete version of multi-objective improved teaching–learning based optimization algorithm (MO-ITLBO) called HTLBO is proposed, employing genetic operators inspired by evolutionary computing methods. Experimental results are extensively compared to those obtained from previous approaches, namely MO-ITLBO, fast elitist non-dominated sorting genetic algorithm (NSGA-II), multi-objective differential evolution (MODE), and multi-objective evolutionary algorithm based on decomposition (MOEA/D). The evaluation shows significant improvements in different metrics, including spacing, hypervolume, non-dominated solutions, and coverage. • We investigate the problem of optimal sensor node placement with three objectives: (i) minimize the number of deployed sensor nodes, (ii) maximize the k -coverage metric, and (iii) maximize the network lifetime. • We propose a hybrid algorithm combining teaching–learning based optimization and evolutionary computing for multi-objective sensor placement. The proposed algorithm introduces multiple teachers to improve the learners' results in different subjects. Moreover, a learner interacts with other learners through a crossover operator. • We compare the proposed algorithm with existing methods, including MODE, MOEAD, MO-TLBO, and NSGA-II on C-metric, spacing-metric, hypervolume-metric, and non-dominated solutions metric to demonstrate its effectiveness. [ABSTRACT FROM AUTHOR]