Control Charts with Variable Dimension for Linear Combination of Poisson Variables

This article analyzes the simultaneous control of several correlated Poisson variables by using the Variable Dimension Linear Combination of Poisson Variables (VDLCP) control chart, which is a variable dimension version of the LCP chart. This control chart uses as test statistic, the linear combination of correlated Poisson variables in an adaptive way, i.e. it monitors either p₁ or p variables (p₁ < p) depending on the last statistic value. To analyze the performance of this chart, we have developed software that finds the best parameters, optimizing the out‐of‐control average run length (ARL) for a shift that the practitioner wishes to detect as quickly as possible, restricted to a fixed value for in‐control ARL. Markov chains and genetic algorithms were used in developing this software. The results show performance improvement compared to the LCP chart. Copyright © 2015 John Wiley & Sons, Ltd.     

Floral nectar phenolics as biochemical markers for the botanical origin of heather honey

In order to find out biochemical markers for the botanical origin of heather (Erica) honey, the phenolic metabolites present in heather floral nectar, collected from the honey-stomach of bees gathering nectar from these flowers, were analysed. The flavonoid fraction of nectar contained four main flavonoids. Two of them were quercetin and kaempferol 3-rhamnosides, and the other two were tentatively identified as myricetin 3-methyl ether and isorhamnetin 3-rhamnosides. Since the natural glycosides are hydrolysed by bee enzymes to render the corresponding aglycones, which are the metabolites detected in honey, acid hydrolysis of the nectar glycosides was achieved. The aglycones quercetin, myricetin 3-methyl ether, kaempferol and isorhamnetin were identified, as well as the gallic acid derivative ellagic acid. The analysis of Portuguese heather honey samples showed that ellagic acid was present in all the samples in significant amounts ranging between 100 g and 600 g per 100 g honey. The other nectar-derived flavonoids were also present, although some of them in very variable amounts. Ellagic acid and myricetin 3-methyl ether, which have not been detected in any of the monofloral honey samples investigated so far, with the only exception being a French honey sample of the botanically relatedCalluna (Ericaceae) which also contained ellagic acid, seem to be the most useful potential markers for the floral origin of heather honey. However, more detailed and extensive investigations are needed to prove the utility of these markers.     

Addressing data complexity for imbalanced data sets: analysis of SMOTE-based oversampling and evolutionary undersampling

In the classification framework there are problems in which the number of examples per class is not equitably distributed, formerly known as imbalanced data sets. This situation is a handicap when trying to identify the minority classes, as the learning algorithms are not usually adapted to such characteristics. An usual approach to deal with the problem of imbalanced data sets is the use of a preprocessing step. In this paper we analyze the usefulness of the data complexity measures in order to evaluate the behavior of undersampling and oversampling methods. Two classical learning methods, C4.5 and PART, are considered over a wide range of imbalanced data sets built from real data. Specifically, oversampling techniques and an evolutionary undersampling one have been selected for the study. We extract behavior patterns from the results in the data complexity space defined by the measures, coding them as intervals. Then, we derive rules from the intervals that describe both good or bad behaviors of C4.5 and PART for the different preprocessing approaches, thus obtaining a complete characterization of the data sets and the differences between the oversampling and undersampling results.     

Tetra-trees properties in graphic interaction

The localization of the components of an object near to a device before obtaining the real interaction is usually determined by means of a proximity measurement to the device of the object’s features. In order to do this efficiently, hierarchical decompositions are used, so that the features of the objects are classified into several types of cells, usually rectangular.In this paper we propose a solution based on the classification of a set of points situated on the device in a little-known spatial decomposition named tetra-tree. Using this type of spatial decomposition gives us several quantitative and qualitative properties that allow us a more realistic and intuitive visual interaction, as well as the possibility of selecting inaccessible components. These features could be used in virtual sculpting or accessibility tasks.In order to show these properties we have compared an interaction system based on tetra-trees to one based on octrees.     

Improved likelihood inference in Birnbaum-Saunders regressions

The Birnbaum-Saunders regression model is commonly used in reliability studies. We address the issue of performing inference in this class of models when the number of observations is small. Our simulation results suggest that the likelihood ratio test tends to be liberal when the sample size is small. We obtain a correction factor which reduces the size distortion of the test. Also, we consider a parametric bootstrap scheme to obtain improved critical values and improved p-values for the likelihood ratio test. The numerical results show that the modified tests are more reliable in finite samples than the usual likelihood ratio test. We also present an empirical application.     

Constrained linear regression models for symbolic interval-valued variables

This paper introduces an approach to fitting a constrained linear regression model to interval-valued data. Each example of the learning set is described by a feature vector for which each feature value is an interval. The new approach fits a constrained linear regression model on the midpoints and range of the interval values assumed by the variables in the learning set. The prediction of the lower and upper boundaries of the interval value of the dependent variable is accomplished from its midpoint and range, which are estimated from the fitted linear regression models applied to the midpoint and range of each interval value of the independent variables. This new method shows the importance of range information in prediction performance as well as the use of inequality constraints to ensure mathematical coherence between the predicted values of the lower () and upper () boundaries of the interval. The authors also propose an expression for the goodness-of-fit measure denominated determination coefficient. The assessment of the proposed prediction method is based on the estimation of the average behavior of the root-mean-square error and square of the correlation coefficient in the framework of a Monte Carlo experiment with different data set configurations. Among other aspects, the synthetic data sets take into account the dependence, or lack thereof, between the midpoint and range of the intervals. The bias produced by the use of inequality constraints over the vector of parameters is also examined in terms of the mean-square error of the parameter estimates. Finally, the approaches proposed in this paper are applied to a real data set and performances are compared.     

Using Scrum to guide the execution of software process improvement in small organizations

For software process improvement - SPI - there are few small organizations using models that guide the management and deployment of their improvement initiatives. This is largely because a lot of these models do not consider the special characteristics of small businesses, nor the appropriate strategies for deploying an SPI initiative in this type of organization. It should also be noted that the models which direct improvement implementation for small settings do not present an explicit process with which to organize and guide the internal work of the employees involved in the implementation of the improvement opportunities. In this paper we propose a lightweight process, which takes into account appropriate strategies for this type of organization. Our proposal, known as a “Lightweight process to incorporate improvements”, uses the philosophy of the Scrum agile method, aiming to give detailed guidelines for supporting the management and performance of the incorporation of improvement opportunities within processes and their putting into practice in small companies. We have applied the proposed process in two small companies by means of the case study research method, and from the initial results, we have observed that it is indeed suitable for small businesses.