Robustness of classification rules that incorporate additional information

The discrimination problem for two normal populations with the same covariance matrix when additional information on the population is available is considered. A study of the robustness properties against training sample contamination of classification rules that incorporate this additional information is performed. These rules have received recently attention where their total misclassification probability (TMP) is proved to be lower than Fisher's linear discriminant rule. The results of a simulation study on the TMP which compares the behaviour of the new rules against Fisher's rule and some of its robustified versions under different types of contamination are presented. These results show that the rules that incorporate the additional information not only have lower TMP, but they also prevent against some types of contamination. In order to achieve prevention from all types of contamination a robustifed version of these rules is recommended.     

Evaluation of the completion of influenza vaccination

OBJECTIVE: To find the reasons which determine failures to comply with anti-flu vaccinations, so that these can be corrected and the coverage of this preventive action be increased. DESIGN: Observational crossover study, done by means of a telephone survey of people over 65. A questionnaire with closed questions, composed after a pilot study and validated by Cronbach's alpha. SETTING: Primary Care Centre (PCC). PATIENTS: We calculated a population sample for qualitative variables (_ = 0.05; p = 0.60; e = 0.05) of 294 people over 65, chosen from the PCC records, by means of random sampling (K = 4) stratified for age and discounting the telephone selection bias. MEASUREMENTS AND RESULTS: The proportion of vaccinated patients (60.9%) obtained in our study did not significantly differ from that in the general population. The percentage of patients included in the programme for the first time was 14%. Level of satisfaction among those vaccinated was 89.4%, with 8.9% of problems detected being light. Main causes of non-vaccination were: thinking that they didn't need it (63.5%), ignorance of the campaign (35.7%), fear of the reaction (24.3%), forgetting (10.4%). The main form of access to the campaign information was from the PCC, both through individuals and posters. Lack of information was statistically significant (p < 0.00001) as a determinant of non-vaccination, without other factors (age, sex, associated pathologies...) explaining these differences. CONCLUSIONS: Individualised and on-going health education by the PCC is fundamental. This would enable the identification of the group not vaccinated due to their express refusal and the recovery of non-vaccinated patients.     

Improving the testability of switched-capacitor filters

A design-for-test methodology for SC filters is presented, based on an architecture using some additional circuitry and providing extra capabilities for both off- and on-line tests. The approach uses a comparison (voting) mechanism to indicate whether or not two copies of a filter element (a biquad, for instance) have a similar response during their actual operation. The design and implementation of a few filter examples are included to assess the potential usefulness of this new approach.     

Preparation of silica by acid dissolution of sepiolite and study of its reinforcing effect in elastomers

The preparation of amorphous silica by acid treatment of a natural magnesium silicate and its use as reinforcing filler for NR and SBR compounds are studied. Limited to the treatment with nitric acid, the rate constants at different temperatures and acid concentrations and the activation energy are calculated, as a function of percentage of extracted magnesium. The resulting material is basically an amorphous silica (92.5% SiO₂), as shown by X-ray analysis, that keeps the fibrous morphology of the original mineral. According to a preliminary technological study the resulting silica is a suitable reinforcing filler for general purpose rubbers; a comparison between the experimental results and those obtained with a commercial fine particle precipitated silica was made. Using a SBR compound the effect of silane coupling agents on the new silica was studied.     

Log-Burr XII regression models with censored data

In survival analysis applications, the failure rate function may frequently present a unimodal shape. In such case, the log-normal or log-logistic distributions are used. In this paper, we shall be concerned only with parametric forms, so a location-scale regression model based on the Burr XII distribution is proposed for modeling data with a unimodal failure rate function as an alternative to the log-logistic regression model. Assuming censored data, we consider a classic analysis, a Bayesian analysis and a jackknife estimator for the parameters of the proposed model. For different parameter settings, sample sizes and censoring percentages, various simulation studies are performed and compared to the performance of the log-logistic and log-Burr XII regression models. Besides, we use sensitivity analysis to detect influential or outlying observations, and residual analysis is used to check the assumptions in the model. Finally, we analyze a real data set under log-Burr XII regression models.     

Linear complete differential resultants and the implicitization of linear DPPEs

The linear complete differential resultant of a finite set of linear ordinary differential polynomials is defined. We study the computation by linear complete differential resultants of the implicit equation of a system of n$n$ linear differential polynomial parametric equations in n−1$n-1$ differential parameters. We give necessary conditions to ensure properness of the system of differential polynomial parametric equations.     

Learning the attachment theory with the CM-ED concept map editor

This paper presents a study carried out at the University of the Basque Country UPV/EHU with the aim of evaluating the CM-ED (concept map editor) with social education students. Concept mapping is a widely accepted technique that promotes meaningful learning. Graphically representing concepts of the learning domain and relationships between them helps students integrate new knowledge into their current cognitive structure. Due to the flexibility of computer-aided drawing graphs, several concept mapping tools have been developed and their use has been studied over the last few years. CM-ED is a multilingual and multimedia software program designed for drawing concept maps. Until recently, CM-ED had been mainly used and evaluated in computer science university degree. This paper represents a qualitative step in the evaluation of CM-ED: from technical students to students of more theoretical fields. The main characteristics of the CM-ED editor and the carried out study are presented in this paper.     

Stochastic automata-based estimators for adaptively compressing files with nonstationary distributions.

This correspondence shows that learning automata techniques, which have been useful in developing weak estimators, can be applied to data compression applications in which the data distributions are nonstationary. The adaptive coding scheme utilizes stochastic learning-based weak estimation techniques to adaptively update the probabilities of the source symbols, and this is done without resorting to either maximum likelihood, Bayesian, or sliding-window methods. The authors have incorporated the estimator in the adaptive Fano coding scheme and in an adaptive entropy-based scheme that "resembles" the well-known arithmetic coding. The empirical results obtained for both of these adaptive methods are obtained on real-life files that possess a fair degree of nonstationarity. From these results, it can be seen that the proposed schemes compress nearly 10% more than their respective adaptive methods that use maximum-likelihood estimator-based estimates.     

Correlation of microhardness and morphology in injection-moulded poly(ethylene terephtalate)

Microindentation hardness has been applied to a series of injection-moulded poly(ethylene terephtalate) samples prepared using a range of mould temperatures, T _c. The morphology of the samples was characterized by X-ray diffraction and differential scanning calorimetry. Depending on T _c, it is shown that microhardness is lower at the surface than in the core of the mouldings. Results are discussed in terms of the volume fraction of spherulites filling the mouldings which is shown to be dependent upon T _c. The influence of an annealing treatment on the properties of the mouldings is examined. The microhardness values are correlated with the thickness and with the surface free energy of the lamellar crystals. The results obtained indicate that increasing annealing temperatures first leads to an increase and then to a sudden decrease of hardness. The latter can be associated with the changes occurring in the number of defects on the crystal's surface.