Transactions on Computational Collective Intelligence XXX

The goal of this paper is to describe a simple protocol based on passive stigmergy for agent interaction in a multiagent system, which can exhibit complex behavior, and to study the effects of alternative decisions, which can be seen as perturbations that can change the final state of the system. Several ways of visualizing the influence relations that the agents have on one another and the effects of alternative decisions are presented.

This research introduces a cognitive computational model of emotional contagion in a crowd of soccer supporters. It is useful for: (1) better understanding of the emotional contagion processes and (2) further development into a predictive and advising application for soccer stadium managers to enhance and improve the ambiance during the soccer game for safety or economic reasons. The model is neurologically grounded and focuses on the emotions “pleasure” and “sadness”. Structured simulations with different crowd compositions and type of matches showed the following four emergent patterns of emotional contagion: (1) hooligans are very impulsive and are not fully open for other emotions, (2) fanatic supporters are very impulsive and open for other emotions, (3) family members are very easily influenced and are not very extravert, (4) the media is less sensitive to the ambiance in the stadium. For model validation, the model outcomes were compared to the heart rate of 100 supporters and reported emotions. The model produced similar heart rate and emotional patterns, thereby establishing its validity. Further implications of the model are discussed.

The wide adoption of the Service Oriented Architecture (SOA) paradigm has provided a means for heterogeneous systems to seamlessly interact and exchange data. Thus, enterprises and end-users have widely utilized Web Services (WS), either as stand-alone applications or as part of more complex service compositions, in order to fulfill their business needs. But, while WS offer a plethora of benefits, a significant challenge rises due to the abundance of available services that can be retrieved online. In this work, we propose a framework for the selection and dynamic composition of WS, by utilizing Linked open Data (LoD). In addition, we propose a hybrid algorithm that uses as input the user’s personalized weights for non-functional characteristics and the results produced by appropriate SPARQL queries that are filtered results using a top-k approach. It then handles the ranking of alternatives based on their population. Finally, using two case studies and a dataset that describes real-world WS, we argue on the feasibility and performance of the proposed method.

In the recent years, IT (Information Technology) offers may have represented a barrier for a customer who does not share the same technical knowledge with providers. Therefore, it would be useful to let a customer express his thoughts and intentions. For this reason, customer’s intentions analysis has become a major contemporary challenge with the relentless growth of the IT market. With an approach for automatically detecting customer’s intention from a free text, it would be possible for a provider to understand the client’s needs and, consequently, the detected intention which may serve as a useful input for recommendation engines. This paper describes an automatic approach that populates an ontology of intentions from client’s textual request in the IT market space. This approach is based on an ontology structure that models the clients’ intentions. It takes an English written request as input and produces an intention ontology instance as output by the means of many combined NLP (Natural Language Processing) techniques. The population process is mainly based on a set of linguistic rules. Moreover, a certainty factor is assigned to each rule serving later as a degree of membership to the concept instantiated with the rule. The empirical evaluation confirms the interesting performance when evaluated on the customers’ requests through a database (available at this link: https://​sites.​google.​com/​view/​customer-request-dataset) specialized in an IT domain.

In this paper, an original enhanced harmony search algorithm (HS-TTP) is proposed for the well-known NP-hard traveling tournament problem (TTP) in sports scheduling. TTP is concerned with finding a tournament schedule that minimizes the total distances traveled by the teams. TTP is well-known, and an important problem within the collective sports communities since a poor optimization in TTP can cause heavy losses in the budget of managing the league’s competition. In order to apply HS to TTP, we use a largest-order-value rule to transform harmonies from real vectors to abstract schedules. We introduce a new heuristic for rearranging the scheduled rounds which give us a significant enhancement in the quality of the solutions. Further, we use a local search as a subroutine in HS to improve its intensification mechanism. The overall method (HS-TTP) is evaluated on publicly available standard benchmarks and compared with the state-of-the-art techniques. Our approach succeeds in finding optimal solutions for several instances. For the other instances, the general deviation from optimality is equal to 4.45%. HS-TTP is able to produce high-quality results compared to existing methods.

Ultrasound imaging is one of the safest and most effective method generally used for the diagnosis of fetal growth. The precise assessment of fetal growth at the time of pregnancy is tough task but ultrasound imaging have improved this vital aspect of Obstetrics and Gynecology. In this paper performance of different learning algorithms of Feed forward neural network based on back-propagation algorithm are analyzed and compared. Basically detection of fetal abnormality using neural network is a hybrid method, in which biometric parameters are extracted and measured from segmentation techniques. Then extracted value of biometric parameters are applied on neural network for detect the fetus status. The artificial neural network (ANN) model is applied for the better diagnosis and effective classification purpose. ANN model are design to discriminate normal and abnormal fetus based on the 2-D US images. In this paper, feed forward back- propagation neural network using Levenberg-Marquardt, Bayesian Regularization and Scaled Conjugate Gradient algorithms are analyzed and used for diagnosis and classification of fetal growth. Performance of these methods are compared and evaluated based on desired output and mean square error. Results found from the Bayesian based neural networks, are in closed confirmation with the real time results. This modeling will help radiologist to take appropriate decision in the boundary line cases.

Many novel challenges and opportunities are associated with Big Data which require rethinking for many aspects of the traditional data warehouse architecture. Indeed, big data are collections of data sets so large and complex to process using classical data warehousing. This data is sourced from many different places such as social media and stored in different formats. It is primarily unstructured data needs a high performance information technology infrastructure that provides superior computational efficiency and storage capacity. This infrastructure should be flexible and scalable to ensure its management over large scale. In recent years, cloud computing is gaining momentum with more and more successful adoptions. This paper proposes a new data warehouse infrastructure as a service to effectively support distribution of big data storage, computing and parallelized programming.

Prolonged climate change contributes to an increase in the local concentrations of O₃ and PMx in the atmosphere, influencing the seasonality and duration of air pollution incidents. Air pollution in modern urban centers such as Athens has a significant impact on human activities such as industry and transport. During recent years the economic crisis has led to the burning of timber products for domestic heating, which adds to the burden of the atmosphere with dangerous pollutants. In addition, the topography of an area in conjunction with the recording of meteorological conditions conducive to atmospheric pollution, act as catalytic factors in increasing the concentrations of primary or secondary pollutants. This paper introduces an innovative hybrid system of predicting air pollutant values (IHAP) using Soft computing techniques. Specifically, Self-Organizing Maps are used to extract hidden knowledge in the raw data of atmospheric recordings and Fuzzy Cognitive Maps are employed to study the conditions and to analyze the factors associated with the problem. The system also forecasts future air pollutant values and their risk level for the urban environment, based on the temperature and rainfall variation as derived from sixteen CMIP5 climate models for the period 2020–2099.

Very often in some censorious healthcare scenario, there may be a need to have some expert consultancies (especially by doctors) that are not available in-house to the hospitals. Earlier, this interesting healthcare scenario of hiring the expert consultants (mainly doctors) from outside of the hospitals had been studied with the robust concepts of mechanism design with money and mechanism design without money. In this paper, we explore the more realistic two sided matching market in our healthcare set-up. In this, the members of the two participating communities, namely the patients and the doctors are revealing the strict preference ordering over the members of the opposite community for a stipulated amount of time. We assume that the patients and doctors are strategic in nature. With the theoretical analysis, we demonstrate that the TOMHECs, that results in the stable allocation of doctors to the patients, satisfies the several economic properties such as strategy-proof-ness (or truthfulness) and optimality. Further, the analytically based analysis of our proposed mechanisms i.e. RAMHECs and TOMHECs are carried out on the ground of the expected distance of the allocation done by the mechanisms from the top most preference. The proposed mechanisms are also validated with the help of exhaustive experiments.

The human drivers in a real world decide and act according to their experience, logic, and judgments. In contrast, mathematical models act according to mathematical equations that ensure the precision of decision to take. However, these models do not provide a promising simulation and they do not reflect the human behaviors. In this context, we present in this paper a completely artificial intelligence anticipation model of car-following problem based on fuzzy logic theory, in order to estimate the velocity of the leader vehicle in near future. The results of experiments, which were conducted by using Next Generation Simulation (NGSIM) dataset to validate the proposed model, indicate that the vehicle trajectories simulated based on the new model are in compliance with the actual vehicle trajectories in terms of deviation and gap distance. In addition, the road security is assured in terms of harmonization between gap distance and security distance.

In this paper we will present one approach and solution for the implementation of load-balancing and fault-tolerance in the XJAF agent middleware. One of the most significant features of this middleware is the use of modern enterprise technologies. Our solution relies on those technologies. First we will briefly present the XJAF architecture and its essential features and functionalities. Then we will compare results of the execution of the same example in two multi-agent frameworks that support clustering: our in-house developed system (XJAF) and widely known and used JADE. We shall demonstrate that a distributed agent application deployed on the XJAF middleware cluster can survive failure of its nodes, while the JADE-based deployment cannot.

Concurrent systems are becoming tremendous in different fields such as network applications, communication protocols and client server applications. However, they are rather difficult to develop and especially, due to concurrency, these systems are faced to specific errors like deadlocks and livelocks. In this context, model checking is a promising formal method which permits systems analysis at early stage, thus ensuring prevention from errors occurring. In previous work [16], we proposed an extension of timed temporal logic TCTL with more powerful modalities aiming to specify properties with clocks quantifiers as well as features for transient states. We formally defined the syntax and the semantics of the proposed quantitative logic called \(TCTL^{\varDelta }_{h}\). As well as in [15], we used timed automata and region graph to discuss the applicability of the proposal to model checking by studying its decidability and complexity.

In this paper, we define a timed temporal logic \(TPN-TCTL^{\varDelta }_{h}\) for time Petri nets, for which model-checking is PSPACE-complete. In fact, Petri nets in general have gained a special interest due to their expressiveness power especially while dealing with concurrency. After detailing the proposed model checking method, we show its development and integration into the tool Romeo. Finally, we prove the efficiency of the method via case studies and simulation results.