Abstract
Inductive Logic Programming (ILP) systems have been largely applied to classification problems with a considerable success. The use of ILP systems in problems requiring numerical reasoning capabilities has been far less successful. Current systems have very limited numerical reasoning capabilities, which limits the range of domains where the ILP paradigm may be applied. This paper proposes improvements in numerical reasoning capabilities of ILP systems. It proposes the use of statistical-based techniques like Model Validation and Model Selection to improve noise handling and it introduces a new search stopping criterium based on the PAG method to evaluate learning performance. We have found these extensions essential to improve on results mer statistical-based algorithms for time series forecasting used in the empirical evaluation study.

Abstract
ILP systems have been largely applied to datamining classification tasks with a considerable success. The use of ILP systems in regression tasks has been far less successful. Current systems have very limited numerical reasoning capabilities, which limits the application of ILP to discovery of functional relationships of numeric nature. This paper proposes improvements in numerical reasoning capabilities of ILP systems for dealing with regression tasks. It proposes the use of statistical-based techniques like Model Validation and Model Selection to improve noise handling and it introduces a new search stopping criterium based on the PAC method to evaluate learning performance. We have found these extensions essential to improve on results over machine learning and statistical-based algorithms used in the empirical evaluation study.

Abstract

Abstract
Relatively simple transformations can speed up the execution of queries for data mining considerably. While some ILP systems use such transformations, relatively little is known about them or how they relate to each other. This paper describes a number of such transformations. Not all of them are novel, but there have been no studies comparing their efficacy. The main contributions of the paper are: (a) it clarifies the relationship between the transformations; (b) it contains an empirical study of what can be gained by applying the transformations; and (c) it provides some guidance on the kinds of problems that are likely to benefit from the transformations.

Abstract
The Watson distribution defined on the hypersphere is much used for modeling axial data. This distribution is rotationally symmetric about the modal axis. Then, in practice, before using the Watson distribution for modeling our data, it is better to test the hypothesis of rotational symmetry. For this purpose, we consider the test given by Prentice. In this paper, we determine the empirical power of this test, when data come from a Watson distribution vs. the alternative, where data come from a mixture of two Watson distributions.

Abstract
Introduction: Self-reported height and weight data have been used in several studies with the purpose of determining the prevalence of overweight and obesity. Despite being a simple methodology, little information exists about the reliability of these measures, namely, in university students. The objective of this study was to determine the sensitivity and specificity of self-reported body mass index (BMI) to evaluate the prevalence of overweight and obesity in university students. Methods: In a convenience sample of 380 university students (226 women and 154 men), weight and height were obtained by self-reported measures and anthropometrie assessment according to international standards methodology (objective). BMI was calculated from self-reported and direct measures. Results: The discrepancy between objective and self-reported weight was not significative. For height, this discrepancy was significantly different in women, in men, and between genders. The difference between BMI values was significantly different in women (0,8 ± 1,1 kg/m2), in men (0,4 ± 1,1 kg /m2) and between genders. Concerning overweight and obesity, according to the objective BMI, the sensitivity was only 50% in women, and 70% in men, while the specificity was 99% in women and 98% in men. Conclusion: Our results show a poor sensitivity of self-reported weight and height data, to estimate overweight and obesity, thus, this method might not be reliable for studies of prevalence of obesity in this population.