Abstract
Recently, meta-learning techniques have been employed to the problem of algorithm recommendation for gene expression data classification. Due to their flexibility, the advice provided to the user was in the form of rankings, which are able to express a preference order of Machine Learning algorithms accordingly to their expected relative performance. Thus, choosing how to learn accurate rankings arises as a key research issue. In this work, the authors empirically evaluated 2 general approaches for ranking prediction and extended them. The results obtained for 49 publicly available microarray datasets indicate that the extensions introduced were very beneficial to the quality of the predicted rankings. © 2010 Springer-Verlag.

Abstract
In large and complex organizations, the development of workflow automation projects is hard. In some cases, a first important step in that direction is the automation of the routing of incoming documents. In this paper, we describe a project to develop a system for the first routing of incoming letters to the right department within a large, public portuguese institution. We followed a data mining approach, where data representing previous routings were analyzed to obtain a model that can be used to route future documents. The approach followed was strongly influenced by some of the limitations imposed by the customer: the budget available was small and the solution should be developed in SQL to facilitate integration with the existing system. The system developed was able to obtain satisfactory results. However, as in any Data Mining project, most of the effort was dedicated to activities other than modelling (e.g., data preparation), which means that there is still plenty of room for improvement.

Abstract
The problem of learning label rankings is receiving increasing attention from several research communities. A number of common learning algorithms have been adapted for this task, including k-Nearest Neighbours (k-NN) and decision trees. Following this line, we propose an adaptation of the naive Bayes classification algorithm for the label ranking problem. Our main idea lies in the use of similarity between the rankings to replace the concept of probability. We empirically test the proposed method on some metalearning problems that consist of relating characteristics of learning problems to the relative performance of learning algorithms. Our method generally performs better than the baseline indicating that it is able to identify some of the underlying patterns in the data.

Abstract
We present a general methodology to model the behavior of heuristics for the Job-Shop Scheduling (JSS) that address the problem by solving conflicts between different operations on the same machine. Our models estimate the gaps between consecutive operations on a machine given measures that characteristics the JSS instance and those operations. These models can be used for a better understanding of the behavior of the heuristics as well as to estimate the performance of the methods. We tested it using two well know heuristics: Shortest Processing Time and Longest Processing Time, that were tested on a large number of random JSS instances. Our results show that it is possible to predict the value of the gaps between consecutive operations from on the job, on random instances. However, the prediction the relative performance of the two heuristics based on those estimates is not successful. Concerning the main goal of this work, we show that the models provide interesting information about the behavior of the heuristics.

Abstract
This paper describes a data mining approach to the problem of detecting erroneous foreign trade transactions in data collected by the Portuguese Institute of Statistics (INE). Erroneous transactions are a minority, but still they have an important impact: on the official statistics produced by INE. Detecting these rare errors is a manual, time-consuming task, which is constrained by a limited amount of available resources (e.g. financial, human). These constraints are common to many other data analysis problems (e.g. fraud detection). Our previous work addresses this issue by producing a ranking of outlyingness that allows a better management of the available resources by allocating them to the most, relevant cases. It is based on an adaptation of hierarchical clustering methods for outlier detection. However, the method cannot be applied to articles with a small number of transactions. In this paper, we complement the previous approach with some standard statistical methods for outlier detection for handling articles with few transactions. Our experiments clearly show its advantages in terms of the criteria, outlined by INE for considering any method applicable to this business problem. The generality of the approach remains to be tested in other problems which share the same constraints (e.g. fraud detection).

Abstract
The values selected for the free parameters of Artificial Neural Networks usually have a high impact on their performance. As a result, several works investigate the use of optimization techniques, mainly metaheuristics, for the selection of values related to the network architecture, like number of hidden neurons, number of hidden layers, activation function, and to the learning algorithm, like learning rate, momentum coefficient, etc. A large number of these works use Genetic Algorithms for parameter optimization. Lately, other bioinspired optimization techniques, like Ant Colony optimization, Particle Swarm Optimization, among others, have been successfully used. Although bioinspired optimization techniques have been successfully adopted to tune neural networks parameter values, little is known about the relation between the quality of the estimates of the fitness of a solution used during the search process and the quality of the solution obtained by the optimization method. In this paper, we describe an empirical study on this issue. To focus our analysis, we restricted the datasets to the domain of gene expression analysis. Our results indicate that, although the computational power saved by using simpler estimation methods can be used to increase the number of solutions tested in the search process, the use of accurate estimates to guide that search is the most important factor to obtain good solutions.