Nowadays, tutoring systems, as well learning content management systems, are widely used in different education environments thanks to their great versatility. Therefore, the challenge of analyzing the large volumes of data generated by them has emerged. In this paper, fuzzy logic techniques are used to predict the students’ academic performance in tutoring systems. With this aim, the use of fuzzy deformable prototypes due to the possibility to adapt the behavior patterns (prototypes) to completely describe a new student is proposed . Moreover, the objective of this research is to extend the capacity of fuzzy deformable prototypes through the use of interval-valued fuzzy sets in order to model the uncertainty of information present in the expert knowledge. Through the use of this enhanced model, satisfactory results have been obtained on the evaluation experiment.

The Curriculum-Based Course Timetabling (CB-CTT) problem involves the task of scheduling lectures of courses to classrooms, considering teacher availability, a specified curricula, and a set of constraints. In the Polytechnic University of Victoria (PUV), Mexico, the CB-CTT problem occurs with the additional constraints that the courses must be assigned to teachers and that a teacher cannot teach more than one course in the same curriculum. This document proposes a methodology to solve this special case CB-CTT problem. The main contributions derived from the research are: (a) the solution of the special case of the CB-CTT problem; (b) the use of a Simulated Annealing based methodology, which relaxes the problem into three simpler subproblems (the assignment, distribution, and scheduling of courses), to reduce the search space and speed up the construction of a solution; and (c) the solution of a real world instance using the proposed methodology. The results show that the approach can construct solutions for the problem in the PUV in a few minutes, and of a similar quality to those manually constructed in a couple of weeks.

Optimization is used in diverse areas of science, technology, and business. Metaheuristics are one of the common approaches for solving optimization problems. In this paper, we propose a novel and functional metaheuristic Fisherman Search Procedure (FSP), to solve global optimization problems, which explores new solutions using a combination of guided and local search. We evaluate the performance of FSP on a set of benchmark functions commonly used for testing global optimization methods. We compare FSP with other heuristic methods referenced in the literature, namely differential evolution (DE), particle swarm optimization (PSO), and greedy randomized adaptive search procedures. Results are analyzed in terms of successful runs, i.e., convergence on global minimum values, and time consumption, demonstrating that FSP can achieve very good performances in most of the cases. In 90 % of the cases, FSP is located among the two better results as for successful runs. On the other hand, with regard to time consumption, FSP shows similar results to PSO and DE, achieving the best and second best results for 82 % of the test functions. Finally, FSP showed to be a simple and robust metaheuristic that achieves good solutions for all evaluated theoretical problems.

This paper presents a charged system search optimization method for finding the optimal cluster centers in a given dataset. CSS algorithm utilizes the Coulomb and Gauss laws from electrostatics to initiate the local search, and Newton second law of motion from mechanics is employed for global search. The efficiency and capability of the proposed algorithm are evaluated on seven datasets and compared with existing $$K$$ -means, GA, PSO and ACO algorithms. From the experimental results, it is found that the proposed algorithm provides more accurate and effective results than other methods being compared.

Swarm intelligence algorithms have been extensively used in clustering-based applications, e.g., image segmentation, which is one of the fundamental components in image analysis and pattern recognition domains. Particle swarm optimization (PSO) is among swarm intelligence algorithms that perform based on population and random search. In this paper, a hybrid algorithm based on PSO, $$k$$ -means, and learning automata is proposed for image segmentation. Each particle in the proposed method has been equipped with a learning automata (LA). In fact, each particle can either update its position by PSO method or select the next position utilizing $$k$$ -means approach in each iteration based on its LA. In other word, the main aim of the hybrid proposed approach was to utilize the efficiency of PSO and $$k$$ -mean methods under supervision of LA. The proposed approach along with other comparative studies has been applied for segmenting standard test images. Efficiency of the proposed method has been compared with that of other methods, and experimental results show the superiority proposed algorithm.

Nowadays, cloud computing is revolutionizing the services provided through the Internet to adapt itself in order to keep the quality of its services. Recent research foresees the advent of a new discipline of agent-based cloud computing systems that can make decisions about adaption in an uncertain environment. This paper discusses the role of argumentation in the next generation of agreement technologies and its use in cloud computing environments.

Policy Reuse is a transfer learning approach to improve a reinforcement learner with guidance from previously learned similar policies. The method uses the past policies as a probabilistic bias where the learner chooses among the exploitation of the ongoing learned policy, the exploration of random unexplored actions, and the exploitation of past policies. In this work, we demonstrate that Policy Reuse further contributes to the learning of the structure of a domain. Interestingly and almost as a side effect, Policy Reuse identifies classes of similar policies revealing a basis of core-policies of the domain. We demonstrate theoretically that, under a set of conditions to be satisfied, reusing such a set of core-policies allows us to bound the minimal expected gain received while learning a new policy. In general, Policy Reuse contributes to the overall goal of lifelong reinforcement learning, as (i) it incrementally builds a policy library; (ii) it provides a mechanism to reuse past policies; and (iii) it learns an abstract domain structure in terms of core-policies of the domain.

The present study is aimed to evaluate levels of air pollution for the Barcelona Metropolitan Region. For this purpose, a newly developed approach called conformal predictors is considered, and, in particular, use is made of the ridge regression confidence machine (RRCM). The hallmark of this method is that it gives valid estimates, i.e. for a given level of significance of prediction, the probability of error does not exceed this level. Moreover, the chosen specification of the RRCM predictor does not place any requirements on data distribution, apart from being independent and identically distributed. A linear ridge regression conformal predictor has been applied to the data. It has allowed to obtain valid interval estimates of annual nitrogen dioxide concentrations with 95 % confidence. The model has provided good results, but to further increase the efficiency of prediction, the RBF kernel has been used. The data for this study have been provided by the XVPCA (Network for Monitoring and Forecasting of Air Pollution) of the Generalitat of Catalonia. The pollutant considered in this paper is nitrogen dioxide. Its values are represented by annual average concentrations within the period from 1998 to 2009. This paper also describes an application of ordinary kriging, and its results have been compared to those of ridge regression conformal predictor.