Boosting the tree augmented naive bayes classifier

The Tree Augmented Naïve Bayes (TAN) classifier relaxes the sweeping independence assumptions of the Naïve Bayes approach by taking account of conditional probabilities. It does this in a limited sense, by incorporating the conditional probability of each attribute given the class and (at most) one other attribute. The method of boosting has previously proven very effective in improving the performance of Naïve Bayes classifiers and in this paper, we investigate its effectiveness on application to the TAN classifier.

Spatial data mining for enhanced soil map modelling

The principle of using induction rules based on spatial environmental data to model a soil map has previously been demonstrated Whilst the general pattern of classes of large spatial extent and those with close association with geology were delineated small classes and the detailed spatial pattern of the map were less well rendered Here we examine several strategies to improve the quality of the soil map models generated by rule induction Terrain attributes that are better suited to landscape description at a resolution of 250 m are introduced as predictors of soil type A map sampling strategy is developed Classification error is reduced by using boosting rather than cross validation to improve the model Further the benefit of incorporating the local spatial context for each environmental variable into the rule induction is examined The best model was achieved by sampling in proportion to the spatial extent of the mapped classes boosting the decision trees and using spatial contextual information extracted from the environmental variables.

Chimeric human papilloma virus-simian/human immunodeficiency virus virus-like-particle vaccines: Immunogenicity and protective efficacy in macaques

Vaccines to efficiently block or limit sexual transmission of both HIV and human papilloma virus (HPV) are urgently needed. Chimeric virus-like-particle (VLP) vaccines consisting of both multimerized HPV L1 proteins and fragments of SIV gag p27, HIV-1 tat, and HIV-1 rev proteins (HPV-SHIV VLPs) were constructed and administered to macaques both systemically and mucosally. An additional group of macaques first received a priming vaccination with DNA vaccines expressing the same SIV and HIV-1 antigens prior to chimeric HPV-SHIV VLP boosting vaccinations. Although HPV L1 antibodies were induced in all immunized macaques, weak antibody or T cell responses to the chimeric SHIV antigens were detected only in animals receiving the DNA prime/HPV-SHIV VLP boost vaccine regimen. Significant but partial protection from a virulent mucosal SHIV challenge was also detected only in the prime/boosted macaques and not in animals receiving the HPV-SHIV VLP vaccines alone, with three of five prime/boosted animals retaining some CD4+ T cells following challenge. Thus, although some immunogenicity and partial protection was observed in non-human primates receiving both DNA and chimeric HPV-SHIV VLP vaccines, significant improvements in vaccine design are required before we can confidently proceed with this approach to clinical trials. (C) 2002 Elsevier Science (USA).

Downregulation of IgE antibody and allergic responses in the lung by epidermal biolistic microparticle delivery

Background: Biolistic injections provide a needle-free delivery of antigen-laden microparticles to the epithelium. The precision of the injection preferentially targets the Langerhans cell network, which, although ideal for vaccination, might not be suitable for the downregulation of immune responses in immunotherapy. Objective: We sought to determine the ability of biolistic injection of antigen into the epithelium of sensitized mice to inhibit IgE antibody and lung inflammatory responses produced by further exposure to antigen. Methods: Mice were sensitized by means of a needle injection of ovalbumin (OVA) in alum and given a series of biolistic injections of OVA or vehicle control, followed by a boost of OVA in alum. Serum IgE and IgG antibodies were measured before and after the boost. The mice were then challenged intranasally, and the infiltration of inflammatory cells was measured by means of bronchoalveolar lavage. Airway reactivity of the challenged mice was measured by examining responses to methacholine with forced oscillatory techniques. Results: Biolistic injection of OVA into the dorsal skin of sensitized mice markedly inhibited IgE and IgG1 antibody responses induced by boosting. IgG2a antibody responses were reduced rather than stimulated. The eosinophilic inflammation in the bronchoalveolar lavage fluid induced by intranasal challenge was also markedly inhibited. Lung hyperreactivity showed an initial increase and then a decrease of responsiveness to methacholine, with elastance returning to the level of unsensitized mice. Biolistic injection into the buccal epithelium was also inhibitory. Conclusions: Biolistic injection of allergen inhibited the boosting of IgE antibody and eosinophilic lung inflammatory responses without inducing TO immunity.