Abstract

Abstract
Several meta-learning approaches have been developed for the problem of algorithm selection. In this context, it is of central importance to collect a sufficient number of datasets to be used as meta-examples in order to provide reliable results. Recently, some proposals to generate datasets have addressed this issue with successful results. These proposals include datasetoids, which is a simple manipulation method to obtain new datasets from existing ones. However, the increase in the number of datasets raises another issue: in order to generate meta-examples for training, it is necessary to estimate the performance of the algorithms on the datasets. This typically requires running all candidate algorithms on all datasets, which is computationally very expensive. In a recent paper, active meta-learning has been used to address this problem. An uncertainty sampling method for the k-NN algorithm using a least confidence score based on a distance measure was employed. Here we extend that work, namely by investigating three hypotheses: 1) is there advantage in using a frequency-based least confidence score over the distance-based score? 2) given that the meta-learning problem used has three classes, is it better to use a margin-based score? and 3) given that datasetoids are expected to contain some noise, are better results achieved by starting the search with all datasets already labeled? Some of the results obtained are unexpected and should be further analyzed. However, they confirm that active meta-learning can significantly reduce the computational cost of meta-learning with potential gains in accuracy.

Abstract
In this paper we examine how the virtual recommendation may influence co-shopping behavior specifically associated with "match dyad" within online social networks. Drawing on better match theory in economics and social network literature, we develop and propose a conceptual model describing how different antecedents of customer's recommendation influence their intentions and leads to positive outcomes, such as recommendation behavior and coshopping influence. We also describe which consumer and network characteristics accentuate this influence examining the moderator effect of customers' incentives, network structure and interactivity. The study contributes to propose that "match dyad" in small social network provide high-quality valuable information that will improve the match between the product and the potential customer and, matched online social connections, act as a proxy for information about the potential market that is difficult and expensive to measure or observe directly such as customer needs, and to access. The paper concludes with a consideration of the implications of predictions for marketing practice.

Abstract
Wafer slicing in photovoltaic industry is mainly done using multi-wire saw machines. The selection of set of bricks (parallelepiped block of crystalline silicon) to be sawn together poses difficult production scheduling decisions. The objective is to maximize the utilization of the available cutting length to improve the process throughput. We address the problem presenting a mathematical formulation and an algorithm that aims to solve it in very short running times while delivering superior solutions. The algorithm employs a reactive greedy randomized adaptive search procedure with some enhancements. Computational experiments proved its effectiveness and efficiency to solve real-world based problems and randomly generated instances. Implementation of an on-line decision system based on this algorithm can help photovoltaic industry to reduce slicing costs making a contribution for its competitiveness against other sources of energy.

Abstract
This paper proposes a methodology to consider the effects of the integration of DG on planning. Since DG has potential to defer investments in networks, the impact of DG on grid capacity is evaluated. A multi-objective optimization tool based on the meta-heuristic MEPSO is used, supporting an alternative approach to exploiting the Pareto front features. Tests were performed in distinct conditions with two well-known distribution networks: IEEE-34 and IEEE-123. The results combined minimization and maximization in order to produce different Pareto fronts and determine the extent of the impact caused by DG. The analysis provides useful information, such as the identification of futures that should be considered in planning. A future means a set of realizations of all uncertainties. MEPSO also presented a satisfactory performance in obtaining the Pareto fronts. © 2011 IEEE.

Abstract
The paper reports on the final results of the European DELLIISS project (www.delliiss.eu) which objective was to establish an Executive Master degree in Innovative Service Systems (EMISS) targeting professional people. The paper explains the systematic process that has been followed for building the content of the EMISS curriculum, including: (1) the collection of the requirements, their prioritization through Think Tanks with professional attendance, and the identification of the associated skills card for this new service systems architect job profile; (2) the elaboration of a knowledge map structuring the components of knowledge available in the ICT, service science, and innovation scientific domains; and finally (3) the definition of the learning trajectories and the associated program content proposed for the EMISS Executive Master, a new diploma that is offered from January 2011 by 6 European Institutions. © 2011 IEEE.