Showing papers on "Ant colony published in 2015"

Integration of graph clustering with ant colony optimization for feature selection

Abstract: A novel supervised filter-based feature selection method using ACO is proposed.Our method integrates graph clustering with a modified ant colony search process.Each feature set is evaluated using a novel measure without using any learning model.The sizes of the final feature set is determined automatically.The method is compared to the state-of-the-art filter and wrapper based methods. Feature selection is an important preprocessing step in machine learning and pattern recognition. The ultimate goal of feature selection is to select a feature subset from the original feature set to increase the performance of learning algorithms. In this paper a novel feature selection method based on the graph clustering approach and ant colony optimization is proposed for classification problems. The proposed method's algorithm works in three steps. In the first step, the entire feature set is represented as a graph. In the second step, the features are divided into several clusters using a community detection algorithm and finally in the third step, a novel search strategy based on the ant colony optimization is developed to select the final subset of features. Moreover the selected subset of each ant is evaluated using a supervised filter based method called novel separability index. Thus the proposed method does not need any learning model and can be classified as a filter based feature selection method. The proposed method integrates the community detection algorithm with a modified ant colony based search process for the feature selection problem. Furthermore, the sizes of the constructed subsets of each ant and also size of the final feature subset are determined automatically. The performance of the proposed method has been compared to those of the state-of-the-art filter and wrapper based feature selection methods on ten benchmark classification problems. The results show that our method has produced consistently better classification accuracies.

A new fuzzy time series forecasting model combined with ant colony optimization and auto-regression

Abstract: This paper presents a new fuzzy time series model combined with ant colony optimization (ACO) and auto-regression. The ACO is adopted to obtain a suitable partition of the universe of discourse to promote the forecasting performance. Furthermore, the auto-regression method is adopted instead of the traditional high-order method to make better use of historical information, which is proved to be more practical. To calculate coefficients of different orders, autocorrelation is used to calculate the initial values and then the Levenberg–Marquardt (LM) algorithm is employed to optimize these coefficients. Actual trading data of Taiwan capitalization weighted stock index is used as benchmark data. Computational results show that the proposed model outperforms other existing models.

Ship's Trajectory Planning for Collision Avoidance at Sea Based on Ant Colony Optimisation

Abstract: Swarm Intelligence (SI) constitutes a rapidly growing area of research. At the same time trajectory planning in a dynamic environment still constitutes a very challenging research problem. This paper presents a new approach to path planning in dynamic environments based on Ant Colony Optimisation (ACO). Assumptions, a concise description of the method developed and results of real navigational situations (case studies with comments) are included. The developed solution can be applied in decision support systems on board a ship or in an intelligent Obstacle Detection and Avoidance system, which constitutes a component of Unmanned Surface Vehicle (USV) Navigation, Guidance and Control systems.

Adaptive directional overcurrent relay coordination using ant colony optimisation

Abstract: The coordination of directional overcurrent relays is most commonly studied based on a fixed network topology within an interconnected power system. Due to its complexity and non-linearity, the relay coordination is formulated as an optimisation problem. Distribution systems often suffer consequences due to the dynamic changes of network topology and operation of elements. Such changes are for example the inputs and outputs of generators, lines and loads. The consequences are reduction of sensitivity and selectivity of relays. The principal objective of this study is to coordinate the directional overcurrent relays based on adaptive protection scheme. The secondary objective is to present the formulation of ant colony algorithm and a comparison of it with the genetic algorithm.

Hybridized ant colony algorithm for the Multi Compartment Vehicle Routing Problem

Abstract: Hybridized ant colony algorithm has been proposed to solve the Multi Compartment Vehicle Routing Problem.Numerical experiments were performed to evaluate the performance of the algorithm.The numerical results showed that the average total length improvement of the proposed HAC over the existing ACS is 5.1%. In addition, the proposed HAC maintains its high performance in large problems on contrary of the existing ACS.The numerical result for the effect of hybridizing the ant colony algorithm with local search schemes has been presented.Illustration of the benefit of using two-compartment vehicles instead of single-compartment vehicles has been presented. Multi Compartment Vehicle Routing Problem is an extension of the classical Capacitated Vehicle Routing Problem where different products are transported together in one vehicle with multiple compartments. Products are stored in different compartments because they cannot be mixed together due to differences in their individual characteristics. The problem is encountered in many industries such as delivery of food and grocery, garbage collection, marine vessels, etc. We propose a hybridized algorithm which combines local search with an existent ant colony algorithm to solve the problem. Computational experiments are performed on new generated benchmark problem instances. An existing ant colony algorithm and the proposed hybridized ant colony algorithm are compared. It was found that the proposed ant colony algorithm gives better results as compared to the existing ant colony algorithm.

Structural link prediction based on ant colony approach in social networks

Abstract: As the size and number of online social networks are increasing day by day, social network analysis has become a popular issue in many branches of science. The link prediction is one of the key rolling issues in the analysis of social network’s evolution. As the size of social networks is increasing, the necessity for scalable link prediction algorithms is being felt more. The aim of this paper is to introduce a new unsupervised structural link prediction algorithm based on the ant colony approach. Recently, ant colony approach has been used for solving some graph problems. Different kinds of networks are used for testing the proposed approach. In some networks, the proposed scalable algorithm has the best result in comparison to other structural unsupervised link prediction algorithms. In order to evaluate the algorithm results, methods like the top- n precision, area under the Receiver Operating Characteristic (ROC) and Precision–Recall curves are carried out on real-world networks.

Ant Colony Extended

Abstract: This paper studies of the performance of Ant Colony Extended to the Travelling Salesman Problem.The algorithm includes two kinds of ants: patrollers (exploration search) and foragers (exploitation search).The algorithm includes a population dynamics allows the algorithm to self-organise.The algorithm is able to solve the Travelling Salesman Problem successfully compared with other classical ant algorithms. Ant Colony Extended (ACE) is a novel algorithm belonging to the general Ant Colony Optimisation (ACO) framework. Two specific features of ACE are: the division of tasks between two kinds of ants, namely patrollers and foragers, and the implementation of a regulation policy to control the number of each kind of ant during the searching process. In addition, ACE does not employ the construction graph usually employed by classical ACO algorithms. Instead, the search is performed using a state space exploration approach. This paper studies the performance of ACE in the context of the Travelling Salesman Problem (TSP), a classical combinatorial optimisation problem. The results are compared with the results of two well known ACO algorithms: ACS and MMAS. ACE shows better performance than ACS and MMAS in almost every TSP tested instance.

Multi-objective ant colony optimization for requirements selection

Abstract: The selection of a set of requirements between all the requirements previously defined by customers is an important process, repeated at the beginning of each development step when an incremental or agile software development approach is adopted. The set of selected requirements will be developed during the actual iteration. This selection problem can be reformulated as a search problem, allowing its treatment with metaheuristic optimization techniques. This paper studies how to apply Ant Colony Optimization algorithms to select requirements. First, we describe this problem formally extending an earlier version of the problem, and introduce a method based on Ant Colony System to find a variety of efficient solutions. The performance achieved by the Ant Colony System is compared with that of Greedy Randomized Adaptive Search Procedure and Non-dominated Sorting Genetic Algorithm, by means of computational experiments carried out on two instances of the problem constructed from data provided by the experts.

‘Lazy’ in nature: ant colony time budgets show high ‘inactivity’ in the field as well as in the lab

Abstract: Social insect colonies are models for complex systems with sophisticated, efficient, and robust allocation of workers to necessary tasks. Despite this, it is commonly reported that many workers appear inactive. Could this be an artifact resulting from the simplified laboratory conditions in most studies? Here, we test whether the time allocated to different behavioral states differs between field and laboratory colonies of Temnothorax rugatulus ants. Our results show no difference in colony time budgets between laboratory and field observations for any of the observed behaviors, including ‘inactivity’. This suggests that, on the timescale of a few months, laboratory conditions do not impact task allocation at the colony level. We thus provide support for a previously untested assumption of laboratory studies on division of labor in ants. High levels of inactivity, common in social insects, thus appear to not be a laboratory artifact, but rather a naturally occurring trait.

New progresses in swarm intelligence-based computation

Abstract: Nature is a great and immense source of inspiration for solving complex problems in the real world. The well-known examples in nature for swarms are bird flocks, fish schools and the colony of social insects. Birds, ants, bees, fireflies, bats, and pigeons are all bringing us various inspirations for swarm intelligence. In 1990s, swarm intelligence algorithms based on ant colony have highly attracted the interest of researchers. During the past two decades, several new algorithms have been developed depending on different intelligent behaviours of natural swarms. This review presents a comprehensive survey of swarm intelligence-based computation algorithms, which are ant colony optimisation, particle swarm optimisation, artificial bee colony, firefly algorithm, bat algorithm, and pigeon inspired optimisation. Future orientations are also discussed thoroughly.

An ant colony based algorithm for overlapping community detection in complex networks

Abstract: Community detection is of great importance to understand the structures and functions of networks. Overlap is a significant feature of networks and overlapping community detection has attracted an increasing attention. Many algorithms have been presented to detect overlapping communities. In this paper, we present an ant colony based overlapping community detection algorithm which mainly includes ants’ location initialization, ants’ movement and post processing phases. An ants’ location initialization strategy is designed to identify initial location of ants and initialize label list stored in each node. During the ants’ movement phase, the entire ants move according to the transition probability matrix, and a new heuristic information computation approach is redefined to measure similarity between two nodes. Every node keeps a label list through the cooperation made by ants until a termination criterion is reached. A post processing phase is executed on the label list to get final overlapping community structure naturally. We illustrate the capability of our algorithm by making experiments on both synthetic networks and real world networks. The results demonstrate that our algorithm will have better performance in finding overlapping communities and overlapping nodes in synthetic datasets and real world datasets comparing with state-of-the-art algorithms.

Availability of food and nesting-sites as regulatory mechanisms for the recovery of ant diversity after fire disturbance

Abstract: Nest-site is an important resource for cavity-nesting ants, what limits colony establishment and structures ant community composition through competition. In ecosystems frequently disturbed by firecontinuous establishment of new colonies is crucial to the process of natural succession. Based on this perspective, we tested the hypothesis that fire reduces the amount of cavities for nesting (e.g., hollow branches, dry leaves curled, and galls), with negative impact on ant biodiversity. We searched for natural cavities and added artificial-nests to assess whether the occupancy rate and its consequences for colony growth. We also evaluated the availability of food sources for ants (EFN plants, honeydew-hemipterans and preys). We found that burned areas had less diverse and structurally simple vegetation. The occupation of natural and artificial nests was the same between the areas, but the reduced availability of nesting-sites in the burned area indicates higher limitation after the fire. This effect was even stronger in foliage habitat compared to the ground. In fact, most of the 11 cavity-nesting species found were typically arboreal. Species richness was lower in burned area, possibly due to lower nesting-sites availability, but the abundance was higher, which may be explained by the greater availability of food resources, mainly EFN-bearing plants. The high food availability may also explain the bigger colony size in burned area, since nectar and honeydew boosts colony growth and low richness prevents competition. In summary, our results show that changes in the availability of nesting sites and food resources may be key mechanisms by which fire changes the ant fauna, specifically cavity-nesting ants in the Brazilian Savanna.

Deadline Constrained Cloud Computing Resources Scheduling through an Ant Colony System Approach

Abstract: Cloud computing resources scheduling is essential for executing workflows in the cloud platform because it relates to both execution time and execution cost. In this paper, we adopt a model that optimizes the execution cost while meeting deadline constraints. In solving this problem, we propose an Improved Ant Colony System (IACS) approach featuring two novel strategies. Firstly, a dynamic heuristic strategy is used to calculate a heuristic value during an evolutionary process by taking the workflow topological structure into consideration. Secondly, a double search strategy is used to initialize the pheromone and calculate the heuristic value according to the execution time at the beginning and to initialize the pheromone and calculate heuristic value according to the execution cost after a feasible solution is found. Therefore, the proposed IACS is adaptive to the search environment and to different objectives. We have conducted extensive experiments based on workflows with different scales and different cloud resources. We compare the result with a particle swarm optimization (PSO) approach and a dynamic objective genetic algorithm (DOGA) approach. Experimental results show that IACS is able to find better solutions with a lower cost than both PSO and DOGA do on various scheduling scales and deadline conditions.

A two-stage ant colony optimization approach based on a directed graph for process planning

Abstract: An innovative approach based on the two-stage ant colony optimization (ACO) approach is used to optimize the process plan with the objective of minimizing total production costs (TPC) against process constraints. First, the process planning (PP) problem is represented as a directed graph that consists of nodes, directed/undirected arcs, and OR relations. The ant colony finds the shortest path on the graph to achieve the optimal solution. Second, a two-stage ACO approach is introduced to deal with the PP problem based on the graph. In the first stage, the ant colony is guided by pheromones and heuristic information of the nodes on the graph, which will be reduced to a simple weighed graph consisting of the favorable nodes and the directed/undirected arcs linking those nodes. In the second stage, the ant colony is guided by heuristic information of nodes and pheromones of arcs on the simple graph to achieve the optimal solution. Third, the simulation experiments for two parts are conducted to illustrate the application of the two-stage ACO approach to the PP problem. The compared results with the results of other algorithms verify the feasibility and competitiveness of the proposed approach.

Threat detection: contextual recognition and response to parasites by ants

Abstract: The ability of an organism to detect threats is fundamental to mounting a successful defense and this is particularly important when resisting parasites. Early detection of parasites allows for initiation of defense mechanisms, which are vital in mitigating the cost of infection and are likely to be especially important in social species, particularly those whose life history makes parasite pressure more significant. However, understanding the relative strength of behavioral responses in different species and situations is still limited. Here, we test the response of individual ants to fungal parasites in 3 different contexts, for 4 ant species with differing life histories. We found that ants from all 4 species were able to detect fungi on their food, environment, and nest mates and initiate avoidance or upregulate grooming behaviors accordingly to minimize the threat to themselves and the colony. Individuals avoided fungal-contaminated surfaces and increased grooming levels in response to fungal-contaminated nest mates. Ants from all species responded qualitatively in a similar way although the species differed quantitatively in some respects that may relate to life-history differences. The results show that ants of multiple species are capable of recognizing fungal threats in various contexts. The recognition of parasite threats may play an important role in enabling ant colonies to deal with the ever-present threat from disease.

Ant Colony Hyper-heuristics for Travelling Salesman Problem☆

Abstract: Metaheuristics are capable of producing good quality solutions. However, they often need to perform some adjustment of relevant parameters in order to be applied to new problems or different problem instances. Furthermore, they often are time-consuming and knowledge intensive procedures that requires deep understanding of the problem domain. This motivates the investigation of developing an algorithm that can produce good quality solutions across different instances and problems and which do not require extensive parameter tuning. Hyper-heuristics is specifically designed to raise the generality of optimisation systems in such a way that the technique can be reused and applied to other different problems. In this work, we develop a generalised heuristic method (hyper-heuristics) based on ant colony algorithms. In our approach, in order to produce a generalized approach, pheromone and visibility values consider a non-domain specific knowledge. Ant colony hyper-heuristics applies the same methodology as ant colony algorithms where it involves two pheromone updating procedures; the local and global update. The global update will use the best solution found at the current iteration to update the pheromone trail and the local update is performed after each ant performs a tour. We apply our work on one routing problem; the Travelling Salesman Problem (TSP). The TSP is a problem of finding the shortest possible tour to visit each city exactly once and it is known to be in the class of NP-hard problems. Results presented in this paper are able to outperform some other popular methods.

Size matters: nest colonization patterns for twig-nesting ants

Abstract: Understanding the drivers of ant diversity and co-occurrence in agroecosystems is fundamental because ants participate in interactions that influence agroecosystem processes. Multiple local and regional factors influence ant community assembly.We examined local factors that influence the structure of a twig-nesting ant community in a coffee system in Mexico using an experimental approach. We investigated whether twig characteristics (nest entrance size and diversity of nest entrance sizes) and nest strata (canopy shade tree or coffee shrub) affected occupation, species richness, and community composition of twig-nesting ants and whether frequency of occupation of ant species varied with particular nest entrance sizes or strata.We conducted our study in a shaded coffee farm in Chiapas, Mexico, between March and June 2012. We studied ant nest colonization by placing artificial nests (bamboo twigs) on coffee shrubs and shade trees either in diverse or uniform treatments. We also examined whether differences in vegetation (no. of trees, canopy cover and coffee density) influenced nest colonization.We found 33 ant species occupying 73% of nests placed. Nest colonization did not differ with nest strata or size. Mean species richness of colonizing ants was significantly higher in the diverse nest size entrance treatment, but did not differ with nest strata. Community composition differed between strata and also between the diverse and uniform size treatments on coffee shrubs, but not on shade trees. Some individual ant species were more frequently found in certain nest strata and in nests with certain entrance sizes.Our results indicate that twig-nesting ants are nest-site limited, quickly occupy artificial nests of many sizes, and that trees or shrubs with twigs of a diversity of entrance sizes likely support higher ant species richness. Further, individual ant species more frequently occupy nests with different sized entrances promoting ant richness on individual coffee plants and trees.

Escape planning in realistic fire scenarios with Ant Colony Optimisation

Abstract: An emergency requiring evacuation is a chaotic event, filled with uncertainties both for the people affected and rescuers. The evacuees are often left to themselves for navigation to the escape area. The chaotic situation increases when predefined escape routes are blocked by a hazard, and there is a need to re-think which escape route is safest. This paper addresses automatically finding the safest escape routes in emergency situations in large buildings or ships with imperfect knowledge of the hazards. The proposed solution, based on Ant Colony Optimisation, suggests a near optimal escape plan for every affected person -- considering dynamic spread of fires, movability impairments caused by the hazards and faulty unreliable data. Special focus in this paper is on empirical tests for the proposed algorithms. This paper brings together the Ant Colony approach with a realistic fire dynamics simulator, and shows that the proposed solution is not only able to outperform comparable alternatives in static and dynamic environments, but also in environments with realistic spreading of fire and smoke causing fatalities. The aim of the solutions is usage by both individuals, such as from a personal smartphone of one of the evacuees, or for emergency personnel trying to assist large groups from remote locations.

Organization of honeydew collection by foragers of different species of ants (Hymenoptera: Formicidae): Effect of colony size and species specificity

Abstract: Aphid honeydew is one of the main energy sources for various ants in the temperate zone, nevertheless relatively little is known about the organization of the work of honeydew foragers (aphid milkers). This study focuses on the honeydew collecting strate- gies used by different ants in steppe and forest multi-species communities in Western Siberia. The behaviour of marked foragers of 12 species (Formica - 7, Lasius - 2, Camponotus - 1, Myrmica - 2) was recorded. Depending on the degree of the aphid milker specializa- tion and degree of protection of the aphids five honeydew collecting strategies of various complexity were distinguished: unspecialized foragers in (I) "unprotected" aphid colonies (attended by ants 95% of time); (III) low "professional" specialization (ants "on duty" constantly attending aphid colonies); (IV) medium and (V) high "professional" specialization (clear division of tasks: honeydew collecting by "shepherds" and protection of trophobionts by "guards"; and honeydew transportation by "transporters" in V). Task specialization of the honeydew foragers is facultative: different ant taxa demonstrate a certain range of the honeydew collecting strategies of different complexities (Formica - I-V, Lasius - I-II, Camponotus - III, Myrmica - I-II) depending on the needs of their colony. The strategy used by ants did not depend on the species of aphid attended, but is strongly dependent on the species of ant, their colony size, available food resources and seasonality. In summer, the aphid milker specialization becomes more complex as ant colony size increases at both intra- and inter-specific levels and when food is scarce. In autumn Formica s. str. ants, which have the most advanced foraging strategy, adopt a simpler honeydew collecting strategy. Overall, the variety of hon- eydew foraging strategies seems to reflect the unequal contribution of dif ferent ants in forming trophobiotic interactions with aphids.

A Proposal of the New owlANT Method for Determining the Distance between Terms in Ontology

Abstract: The article presents the owlANT method, which allows us to associate a collection of short text messages with ontology The trails were conducted on the collection of communications by Reuters (the so-called second collection) As the methodological base of the method, a swarm intelligence was used, namely the ant colony optimisation The ants, moving between the ontological nodes [1][2], left their pheromone trace As a result, some branches of relations - after some time in evolution - were marked more strongly than the others On the basis of the intensity of the pheromone trace, one can formulate the strength of relations between the various associations, and - indirectly - also between the associated documents As far as the authors know, no one has so far published research on the application of ACO to the development of similarity measure of text documents with the consideration of their meaning-related embedding in ontology The authors refer to the work [3][4], in which the ACO was used for aggregation of concepts in ontology; however, both the purpose and the method were different in that case The research described in the report will be continued to specify the similarity measures taking into consideration the distance in ontology obtained thanks to the evolutionary processing of the meaning of terms with the use of ACO

Branching angles reflect a trade‐off between reducing trail maintenance costs or travel distances in leaf‐cutting ants

Abstract: The design of transport paths in consuming entities that use routes to access food should be under strong selective pressures to reduce costs and increase benefits. We studied the adaptive nature of branching angles in foraging trail networks of the two most abundant tropical leaf-cutting ant species. We mathematically assessed how these angles should reflect the relative weight of the pressure for reducing either trail maintenance effort or traveling distances. Bifurcation angles of ant foraging trails strongly differed depending on the location of the nests. Ant colonies in open areas showed more acute branching angles, which best shorten travel distances but create longer new trail sections to maintain than a perpendicular branch, suggesting that trail maintenance costs are smaller compared to the benefit of reduced traveling distance. Conversely, ant colonies in forest showed less acute branching angles, indicating that maintenance costs are of larger importance relative to the benefits of shortening ...

Multiple Boosting in the Ant Colony Decision Forest meta-classifier

Abstract: The idea of ensemble methodology is to combine multiple predictive models in order to achieve a better prediction performance In this task we analyze the self-adaptive methods for improving the performance of Ant Colony Decision Tree and Forest algorithms Our goal is to present and compare new meta-ensemble approaches based on Ant Colony Optimization The proposed meta-classifiers (consisting of homogeneous classifiers) can be characterized by the self-adaptability or the good accommodation with the analyzed data sets and offer appropriate classification accuracyIn this article we provide an overview of ensemble methods in classification tasks and concentrate on the different methodologies, such as Bagging, Boosting and Random Forest We present all important types of ensemble methods including Boosting and Bagging in context of distributed approach, where agent-ants create better solutions employing adaptive mechanisms Self adaptive, combining methods and modeling appropriate issues, such as ensembles presented here are discussed in context of the quality of the results Smaller trees in decision forest without loss of accuracy are achieved during the analysis of different data sets

The Pied Piper: A Parasitic Beetle’s Melodies Modulate Ant Behaviours

Abstract: Ants use various communication channels to regulate their social organisation. The main channel that drives almost all the ants’ activities and behaviours is the chemical one, but it is long acknowledged that the acoustic channel also plays an important role. However, very little is known regarding exploitation of the acoustical channel by myrmecophile parasites to infiltrate the ant society. Among social parasites, the ant nest beetles (Paussus) are obligate myrmecophiles able to move throughout the colony at will and prey on the ants, surprisingly never eliciting aggression from the colonies. It has been recently postulated that stridulatory organs in Paussus might be evolved as an acoustic mechanism to interact with ants. Here, we survey the role of acoustic signals employed in the Paussus beetle-Pheidole ant system. Ants parasitised by Paussus beetles produce caste-specific stridulations. We found that Paussus can “speak” three different “languages”, each similar to sounds produced by different ant castes (workers, soldiers, queen). Playback experiments were used to test how host ants respond to the sounds emitted by Paussus. Our data suggest that, by mimicking the stridulations of the queen, Paussus is able to dupe the workers of its host and to be treated as royalty. This is the first report of acoustic mimicry in a beetle parasite of ants.