Wyniki wyszukiwania - Biblioteka Nauki

1

Aggregation Pheromone Density Based Pattern Classification

100%

Halder A. , Ghosh A. , Ghosh S.

|

tom Vol. 92, nr 4

345-362

EN

The study of ant colonies behavior and their self-organizing capabilities is of interest to machine learning community, because it provides models of distributed adaptive organization which are useful to solve difficult optimization and classification problems among others. Social insects like ants, bees deposit pheromone (a type of chemical) in order to communicate between the members of their community. Pheromone, that causes clumping behavior in a species and brings individuals into a closer proximity, is called aggregation pheromone. This article presents a new algorithm (called, APC) for pattern classification based on this property of aggregation pheromone found in natural behavior of real ants. Here each data pattern is considered as an ant, and the training patterns (ants) form several groups or colonies depending on the number of classes present in the data set. A new test pattern (ant) will move along the direction where average aggregation pheromone density (at the location of the new ant) formed due to each colony of ants is higher and hence eventually it will join that colony. Thus each individual test pattern (ant) will finally join a particular colony. The proposed algorithm is evaluated with a number of benchmark data sets as well as various kinds of artificially generated data sets using three evaluationmeasures. Results are compared with four other well known conventional classification techniques. Experimental results show the potentiality of the proposed algorithm in terms of all the evaluation measures compared to other algorithms.

2

Swarm intelligence for network routing optimization

100%

Dempsey P. , Schuster A.

Journal of Telecommunications and Information Technology

|

2005

|

tom nr 3

24-28

EN

This paper presents the results of a comparative study of network routing approaches. Recent advances in the field suggest that swarm intelligence may offer a robust, high quality solution. The overall aim of the study was to develop a framework to facilitate the empirical evaluation of a swarm intelligence routing approach compared to a conventional static and dynamic routing approach. This paper presents a framework for the simulation of computer networks, collection of performance statistics, generation and reuse of network topologies and traffic patterns.

3

Swarm intelligence algorithm based on competitive predators with dynamic virtual teams

100%

Yang S. , Sato Y.

Journal of Artificial Intelligence and Soft Computing Research

|

2017

|

tom Vol. 7, No. 2

87--101

EN

In our previous work, Fitness Predator Optimizer (FPO) is proposed to avoid premature convergence for multimodal problems. In FPO, all of the particles are seen as predators. Only the competitive, powerful predator that are selected as an elite could achieve the limited opportunity to update. The elite generation with roulette wheel selection could increase individual independence and reduce rapid social collaboration. Experimental results show that FPO is able to provide excellent performance of global exploration and local minima avoidance simultaneously. However, to the higher dimensionality of multimodal problem, the slow convergence speed becomes the bottleneck of FPO. A dynamic team model is utilized in FPO, named DFPO to accelerate the early convergence rate. In this paper, DFPO is more precisely described and its variant, DFPO-r is proposed to improve the performance of DFPO. A method of team size selection is proposed in DFPO-r to increase population diversity. The population diversity is one of the most important factors that determines the performance of the optimization algorithm. A higher degree of population diversity is able to help DFPO-r alleviate a premature convergence. The strategy of selection is to choose team size according to the higher degree of population diversity. Ten well-known multimodal benchmark functions are used to evaluate the solution capability of DFPO and DFPO-r. Six benchmark functions are extensively set to 100 dimensions to investigate the performance of DFPO and DFPO-r compared with LBest PSO, Dolphin Partner Optimization and FPO. Experimental results show that both DFPO and DFPO-r could demonstrate the desirable performance. Furthermore, DFPO-r shows better robustness performance compared with DFPO in experimental study.

4

Further improvement of the Glowworm Swarm Optimization algorithm by adding a conservation of agent move direction

100%

Oramus P.

|

2011

|

tom Vol. 7, no. 4

15--21

EN

Glowworm Swarm Optimization is an algorithm, which can localize multiple optima of the multi-modal function during a single run. Unfortunately, original algorithm manifests several disadvantages: an agent can change its position only in the presence of other agents and exclusively in direction of randomly chosen neighboring agent. This paper shows how GSO algorithm can be significantly improved by simple modifications: agents receive an alternative method of changing the position by random jumps and they prefer directions consistent with the direction of the previous movement.

5

The Swarm Computing Approach to Business Intelligence

100%

Schumann A. , Pancerz K. , Szelc A.

|

nr 3

41-50

EN

We have proposed to use some features of swarm behaviours in modelling business processes. Due to these features we deal with a propagation of business processes in all accessible directions. This propagation is involved into our formalization instead of communicating sequential processes. As a result, we have constructed a business process diagram language based on the swarm behavior and an extension of that language in the form of reflexive management language.

6

A comparative study on multi-swarm optimisation and bat algorithm for unconstrained non linear optimisation problems

88%

Baidoo E. , Opoku Oppong S.

Applied Computer Science

|

2016

|

tom Vol. 12, no 4

59--77

EN

A study branch that mocks-up a population of network of swarms or agents with the ability to self-organise is Swarm intelligence. In spite of the huge amount of work that has been done in this area in both theoretically and empirically and the greater success that has been attained in several aspects, it is still ongoing and at its infant stage. An immune system, a cloud of bats, or a flock of birds are distinctive examples of a swarm system. In this study, two types of meta-heuristics algorithms based on population and swarm intelligence - Multi Swarm Optimization (MSO) and Bat algorithms (BA) – are set up to find optimal solutions of continuous non-linear optimisation models. In order to analyze and compare perfect solutions at the expense of performance of both algorithms, a chain of computational experiments on six generally used test functions for assessing the accuracy and the performance of algorithms, in swarm intelligence fields are used. Computational experiments show that MSO algorithm seems much superior to BA.

7

From Egyptian barque oracles to Artificial Swarm Intelligence via the Ouija (or wDA?) board

88%

Graham L. D.

|

nr 31

7-41

EN

Ancient Egyptian barque oracles had a recent counterpart in the phenomenon of “table turning”, an occult process experienced in Nineteenth Century Spiritualist séances. The séance table’s small scale successor, the Talking Board, ensured that oracular locomotion persisted throughout the Twentieth Century; its best known embodiment – the Ouija board – remains popular even today. Scientific studies have helped to elucidate the behavioural drivers that govern table turning and Ouija sessions; these reveal that good faith groups are dominated by an auto suggestive process known as the ideomotor response (IMR). Learnings from such studies suggest that a hierarchy of up to four drivers (two conscious and conditional, two unconscious and continuous) would have underpinned the significance laden movements of Egyptian barque oracles. Such oracles constitute an early form of what is today called Artificial Swarm Intelligence (ASI); “human swarming” enables networks of individuals whose interactions are governed by real time feedback loops to converge quickly on optimal solutions. The paper also examines the possibility that Ouija – the name bestowed in 1891 upon the “Egyptian luck board” that went on to dominate the Talking Board market – might genuinely reflect an ancient Egyptian word with the approximate sense of “good luck”, just as the board’s pioneers claimed it did.

8

Improvements to Glowworm Swarm Optimization algorithm

88%

Oramus P.

Computer Science

|

2010

|

tom Vol. 11

7-20

EN

Glowworm Swarm Optimization algorithm is applied for the simultaneous capture of multiple optima of multimodal functions. The algorithm uses an ensemble of agents, which scan the search space and exchange information concerning a fitness of their current position. The fitness is represented by a level of a luminescent quantity called luciferin. An agent moves in direction of randomly chosen neighbour, which broadcasts higher value of the luciferin. Unfortunately, in the absence of neighbours, the agent does not move at all. This is an unwelcome feature, because it diminishes the performance of the algorithm. Additionally, in the case of parallel processing, this feature can lead to unbalanced loads. This paper presents simple modifications of the original algorithm, which improve performance of the algorithm by limiting situations, in which the agent cannot move. The paper provides results of comparison of an original and modified algorithms calculated for several multimodal test functions.

PL

Algorytm Glowworm Swarm Optimization jest stosowany do równoczesnego odnajdywania wielu optimów funkcji multimodalnych. Algorytm używa zespołu agentów przeszukujących przestrzeń poszukiwań i wymieniających się informacjami o wartości funkcji przystosowania w danym położeniu. Funkcja przystosowania jest reprezentowana przez poziom emitującego światło pigmentu - lucyferyny. Agenci poruszają się w kierunku losowo wybranego sąsiada, który rozgłasza wyższą wartość poziomu lucyferyny. Niestety w przypadku braku sąsiadów agent nie porusza się wcale. Stanowi to niepożądaną cechę algorytmu ograniczającą jego wydajność. W przypadku przetwarzania równoległego cecha ta może prowadzić do niezrównoważenia obciążenia. Praca ta przedstawia proste modyfikacje oryginalnego algorytmu zwiększające jego wydajność poprzez ograniczanie liczby takich sytuacji, w których agent nie może się poruszyć. Przedstawione zostały wyniki porównania pracy oryginalnego i zmodyfikowanych algorytmów dla kilku funkcji testowych.

9

An application of swarm intelligence binary particle swarm optimization (BPSO) algorithm to multi-focus image fusion

88%

Zhang X. , Sun L. , Han J. , Chen G.

|

tom Vol. 40, nr 4

949--964

EN

In this paper, an optimal and intelligent multi-focus image fusion algorithm is presented, expected to achieve perfect reconstruction or optimal fusion of multi-focus images with high speed. A synergistic combination of segmentation techniques and binary particle swarm optimization (BPSO) intelligent search strategies is employed in salience analysis of contrast feature-vision system. Also, several evaluations concerning image definition are exploited and used to evaluate the performance of the method proposed. Experiments are performed on a large number of images and the results show that the BPSO algorithm is much faster than the traditional genetic algorithm. The method proposed is also compared with some classical or new fusion methods, such as discrete wavelet-based transform (DWT), nonsubsampled contourlet transform (NSCT), NSCT-PCNN (pulse coupled neural networks (PCNN) method in NSCT domain) and curvelet transform. The simulation results with high accuracy and high speed prove the superiority and effectiveness of the present method.

10

Integrating Forms of Interaction in a Distributed Model

88%

Bandini S. , Simone C.

|

2004

|

tom Vol. 61, nr 1

1--17

EN

The main aim of this paper is to present the formal description of the Multilayered Reaction-Diffusion Machine (MRDM, an extension of a previously introduced RDM, Reaction-Diffusion Machine) which can be seen as a generalization of Cellular Automata since it relaxes some constraints on uniformity, locality and closure. The MRDM offers a formal and computational environment where to describe, represent and simulate coordination models which explicitly require spatial features to be considered and integrates different forms of interaction. The paper is divided into two main parts. In the first one, the formal description of the MRDM is presented. In the second part the MRDM is put in relation with deterministic Turing machines; moreover we show how the MRDM, under determined constraints, collapses on a traditional CA.

11

Fireworks Algorithm for Unconstrained Function Optimization Problems

88%

Baidoo E.

|

tom Vol. 13, no 1

61--74

EN

Modern real world science and engineering problems can be classified as multi-objective optimisation problems which demand for expedient and efficient stochastic algorithms to respond to the optimization needs. This paper presents an object-oriented software application that implements a firework optimization algorithm for function optimization problems. The algorithm, a kind of parallel diffuse optimization algorithm is based on the explosive phenomenon of fireworks. The algorithm presented promising results when compared to other population or iterative based meta-heuristic algorithm after it was experimented on five standard ben-chmark problems. The software application was implemented in Java with interactive interface which allow for easy modification and extended expe-rimentation. Additionally, this paper validates the effect of runtime on the al-gorithm performance.

12

Intelligent algorithms for routing sensory networks operating in explosion hazard zones

88%

Stankiewicz K. , Jagoda J. , Tonkins M.

Mining Science

|

2021

|

tom Vol. 28

103--115

EN

The article presents intelligent routing algorithms currently used in sensory networks, in terms of determining the possibility of their integration into systems working in potentially explosive atmospheres. Selected types of scribing algorithms were characterized. The analysis of simulation tests performed on selected types of scribing algorithms was carried out. The analysis of equipment solutions which can be used to build a network node operating in the conditions of methane and/or coal dust explosion hazard was carried out.

13

Hybridization processes in putative hybrid swarms of scots pine and mountain dwarf pine as revealed by chloroplast DNA

75%

Kormutak A. , Brana M. , Manka P. , Galgoci M. , Libantova J. , Camek V. , Bolecek P. , Gomory D.

|

nr 2

14

FLC control for tuning exploration phase in bio-inspired metaheuristic

75%

Kiełkowicz K. , Grela D.

Annales Universitatis Mariae Curie-Skłodowska. Sectio AI, Informatica

|

2016

|

tom Vol. 16, no. 2

32--38

EN

Growing popularity of the Bat Algorithm has encouraged researchers to focus their work on its further improvements. Most work has been done within the area of hybridization of Bat Algorithm with other metaheuristics or local search methods. Unfortunately, most of these modifications not only improves the quality of obtained solutions, but also increases the number of control parameters that are needed to be set in order to obtain solutions of expected quality. This makes such solutions quite impractical. What more, there is no clear indication what these parameters do in term of a search process. In this paper authors are trying to incorporate Mamdani type Fuzzy Logic Controller (FLC) to tackle some of these mentioned shortcomings by using the FLC to control the exploration phase of a bio-inspired metaheuristic. FLC also allows us to incorporate expert knowledge about the problem at hand and define expected behaviors of system – here process of searching in multidimensional search space by modeling the process of bats hunting for their prey.

15

Optymalizacja przy pomocy roju cząstek bazy reguł klasyfikatora rozmytego

75%

Głuszek A. , Gorzałczany M. B. , Rudziński F.

Logistyka

|

2014

|

tom nr 6

3960--3968

PL

Systemy rozmyte wykorzystywane są często jako systemy eksperckie, w tym systemy klasyfikacji danych. Szczególnie ważną kwestią jest w tym przypadku stworzenie i optymalizacja bazy reguł rozmytych na podstawie danych, opisujących konkretne zagadnienie. W tym celu stosowane są głównie metody z obszaru tzw. inteligencji obliczeniowej (ang. computational intelligence), zwłaszcza z zakresu algorytmów ewolucyjnych. Pierwsza część niniejszego artykułu prezentuje zastosowanie, należącego do obszaru tzw. inteligencji rojowej, algorytmu optymalizacji rojem cząstek PSO (ang. Particle Swarm Optimization) do optymalizacji bazy reguł klasyfikatora rozmytego. W drugiej części artykułu przedstawiono zastosowanie proponowanego podejścia do problemu klasyfikacji dwóch zbiorów danych, pochodzących ze znanej bazy UCI Machine Learning Repository (tzw. Iris Data i Glass Identification Data). Uzyskane wyniki porównano z rezultatami działania metod alternatywnych  algorytmu największego spadku oraz klasycznego algorytmu genetycznego.

EN

Fuzzy systems are often used as expert systems including data classification systems. Very important issue is designing and optimization of their fuzzy rule bases from available pattern data. For this purpose, methods from the area of computational intelligence (especially evolutionary algorithms) are usually applied. The first part of this paper presents an application of the particle swarm optimization algorithm for optimizing rule base of the fuzzy classifier. Particle swarm optimization belongs to the class of swarm intelligence algorithms. Swarm intelligence techniques use the collective behaviors of many simple agents which interact with each other and with their environment. The second part of the paper describes the application of the presented method for the classification of two well-known data sets from UCI Machine Learning Repository (Iris Data and Glass Identification Data). The proposed approach is also compared with two alternative methods - gradient descent and genetic algorithm

16

Dispersed, self-organizing sensory networks supporting the technological

75%

Jagoda J. , Hetmańczyk M. , Stankiewicz K.

Mining Machines

|

2021

|

tom no. 2

13--23

EN

Examples of automation of technological processes for mineral extraction are presented. Aspects related to the diagnostics of machines and devices during operation processes are discussed. Applying the distributed sensor networks to enable designing and manufacturing the machines and devices operated in accordance with the idea of Industry 4.0, the Internet of Things, M2M communication and autonomous behaviour was proposed. The paper presents impact of applying the distributed sensor networks on increasing work safety (multi-redundant communication) and reducing the employment in hazardous areas is presented. Implementation of algorithms based on swarm intelligence to control the routing processes of distributed sensor networks was suggested. Areas of application of distributed sensor networks based on swarm intelligence in other industries (renewable energy sources) are also outlined.

17

Ant colony optimization in project management

75%

Skołud B. , Marcińczyk B.

Computer Assisted Mechanics and Engineering Sciences

|

2007

|

tom Vol. 14, No. 4

745-752

EN

This paper presents an Ant Colony Optimization (ACO) approach to the resource-constrained project scheduling problem (RCPSP). RCPSP as a generalization of the classical job shop scheduling problem belongs to the class of NP-hard optimization problems. Therefore, the use of heuristic solution procedures when solving large problem is well-founded. Most of the heuristic methods used for solving resource-constrained project scheduling problems either belong to the class of priority rule based methods or to the class of metaheuristic based approaches. ACO is a metaheuristic method in which artificial ants build solutions by probabilistic selecting from problem-specific solutions components influenced by a parametrized model of solution, called pheromone model. In ACO several generations of artificial ants search for good solution. Every ant builds a solution step by step going through several probabilistic decisions. If ant find a good solution mark their paths by putting some amount of pheromone (which is guided by some problem specific heuristic) on the edges of the path.

18

Register Cellular Automata in the Hyperbolic Plane

75%

Grigorieff S. , Margenstern M.

|

2004

|

tom Vol. 61, nr 1

19--27

EN

In this paper, we consider cellular automata on special grids of the hyperbolic plane: the grids are constructed on infinigons, i.e. polygons with infinitely many sides. We show that the truth of arithmetical formulas can be decided in finite time with infinite initial recursive configurations. Next, we define a new kind of cellular automata, endowed with data and more powerful operations which we call register cellular automata. This time, starting from finite configurations, it is possible to decide the truth of arithmetic formulas in linear time with respect to the size of the formula.

19

Application of swarm intelligence algorithms in control problems

63%

Kacerka J.

|

2009

|

tom z. 117

13-21

EN

An application of swarm intelligence algorithms to control problems which may be described as an optimisation task is analysed in the paper. A multi-purpose implementation of PSO algorithm allowing for optimisation requiring a reduced number of fitness function evaluations is presented. The conducted experiments show the effectiveness and efficiency of the proposed algorithms.

PL

W pracy przedstawiono analizę możliwości zastosowania algorytmów wykorzystujących inteligencję roju w problemach sterowania, które sprowadzić można do zadania optymalizacji. Opracowano uniwersalną implementację algorytmu PSO pozwalającą na przeprowadzenie skutecznej optymalizacji przy zmniejszonej ilości wykonywanych obliczeń funkcji celu. Proponowane rozwiązania obejmują adaptacyjną zmianę parametrów roju przy pomocy układu logiki rozmytej oraz budowę modelu funkcji celu na podstawie punktów uzyskanych we wcześniejszych iteracjach. Przedstawiono sposoby przystosowania algorytmu PSO do rozwiązania zadania identyfikacji parametrycznej modeli nieliniowych oraz strojenia regulatorów (gdzie zaproponowano ekstrapolację optymalizowanego kryterium po wykonaniu niepełnej symulacji lub pomiaru). Skuteczność proponowanego rozwiązania zbadano na stanowisku badawczym wyposażonym w procesor sygnałowy.

20

A nature inspired collision avoidance algorithm for ships

63%

Lazarowska A.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2023

|

tom Vol. 17 No. 2

341--346

EN

Nature inspired algorithms are regarded as a powerful tool for solving real life problems. They do not guarantee to find the globally optimal solution, but can find a suboptimal, robust solution with an acceptable computational cost. The paper introduces an approach to the development of collision avoidance algorithms for ships based on the firefly algorithm, classified to the swarm intelligence methods. Such algorithms are inspired by the swarming behaviour of animals, such as e.g. birds, fish, ants, bees, fireflies. The description of the developed algorithm is followed by the presentation of simulation results, which show, that it might be regarded as an efficient method of solving the collision avoidance problem. Such algorithm is intended for use in the Decision Support System or in the Collision Avoidance Module of the Autonomous Navigation System for Maritime Autonomous Surface Ships.