On boosting kernel density methods for multivariate data: density estimation and classification

Statistical learning is emerging as a promising field where a number of algorithms from machine learning are interpreted as statistical methods and vice-versa. Due to good practical performance, boosting is one of the most studied machine learning techniques. We propose algorithms for multivariate density estimation and classification. They are generated by using the traditional kernel techniques as weak learners in boosting algorithms. Our algorithms take the form of multistep estimators, whose first step is a standard kernel method. Some strategies for bandwidth selection are also discussed with regard both to the standard kernel density classification problem, and to our 'boosted' kernel methods. Extensive experiments, using real and simulated data, show an encouraging practical relevance of the findings. Standard kernel methods are often outperformed by the first boosting iterations and in correspondence of several bandwidth values. In addition, the practical effectiveness of our classification algorithm is confirmed by a comparative study on two real datasets, the competitors being trees including AdaBoosting with trees.     

Adolescent Depression as a Contributing Factor to the Development of Substance Use Disorders

An exploratory investigation of the variables linking adolescent substance abuse and depression was conducted using the “grounded theory” qualitative research approach. A convenience sample was drawn from African American adolescents, aged 12 to 18 years, who reside in a public housing community in Baltimore, Maryland. The results revealed approximately 5% of the sample openly admitted to using drugs because they are sad, feel like a failure, lack energy and because they have family problems. Moreover, 26% of the respondents reported drinking alcoholic beverages; 26% of the population reported smoking marijuana and 16% of the respondents reported they smoke cigarettes. Additionally, life stressors (i.e. an absent parent from the home, childhood illnesses, family discord, living in kinship or foster care environments, living in single-parent households and unemployment) are adversely impacting the lives of African American adolescents residing in a public housing community.     

Ethics and social media: Implications for sociolinguistics in the networked public1

As a popular agora for writing identity into being, the networked public of social media sites presents exciting and unprecedented possibilities for sociolinguistic research. At the same time, these sites raise a wealth of unfamiliar methodological and ethical issues, and debate concerning appropriate ethical measures for research targeting online discourse communities is emergent. One of the most pressing debates concerns the visibility of online interaction (i.e. its locus between the public and private ends of the continuum). Although they exist freely online, networked publics are not public forums. They are governed by both personal and communal norms, and they are networked. This combination of factors gives rise to unique ethical challenges, particularly in the case of Facebook, an accessible and data‐rich, yet problematic, research site. This paper reviews the ethical difficulties presented by Facebook, and presents a framework for ethnographic sociolinguistic research that uses this site as a source of data. En tant qu’espace public privilégié pour la création de l’identitéà travers l’écriture, le public ?réseauté? (interconnecté) des sites des médias sociaux présente des possibilités prometteuses et sans précédent pour la recherché en sociolinguistique. Cependant, ces sites soulèvent de nombreuses questions méthodologiques et éthiques qui sont nouvelles, d’où l’émergence de discussions portant sur les mesures éthiques appropriées pour la recherche qui vise les communautés discursives en ligne. Un des débats concerne la visibilité de l’interaction en ligne (c.‐à‐d. sa place entre les extrémités privée et publique du continuum). Bien qu’ils existent de façon libre en ligne, les réseaux publics ne sont pas des forums publics. Ils sont gouvernés à la fois par des normes personnelles et communautaires, et ils sont interconnectés. Cette combinaison de facteurs donne naissance à des défis éthiques uniques, en particulier dans le cas de Facebook, un site de recherche accessible et riche en données, mais problématique. Cet article passe en revue les difficultés éthiques que Facebook présente, et offre un cadre de travail pour la recherche sociolinguistique ethnographique utilisant ce site comme source de données. [French]     

Tourist profiles using discriminant analysis

How similar are tourists from the audiences of two or more advertising vehicles? The method of N-way multiple discriminant analysis was used for predicting magazine readership from key tourist variables from two samples of 300 coupon inquiries. Results of the analysis indicated similar tourist profiles existed among the readerships of the three magazines. Application of the method used in this study may be useful in the development of promotional themes to attract tourists from selected advertising vehicles and in identifying unique tourism characteristics for particular vehicles.     

Procrustes shape analysis of triangulations of a two coloured point pattern

Consider a set of points in the plane with Gaussian perturbations about a regular mean configuration in which a Delaunay triangulation of the mean of the process is comprised of equilateral triangles of the same size. The points are labelled at random as black or white with variances of the perturbations possibly dependent on the colour. By investigating triangle subsets (with four sets of possible colour labels for the vertices) in detail we propose various test statistics based on a Procrustes shape analysis. A simulation study is carried out to investigate the relative merits and the adequacy of the approximations used in the distributional results, as well as a comparison with simulation methods based on nearest-neighbour distances. The methodology is applied to an investigation of regularity in human muscle fibre cross-sections.     

Block diagrams and splitting criteria for classification trees

Various aspects of the classification tree methodology of Breiman et al., (1984) are discussed. A method of displaying classification trees, called block diagrams, is developed. Block diagrams give a clear presentation of the classification, and are useful both to point out features of the particular data set under consideration and also to highlight deficiencies in the classification method being used. Various splitting criteria are discussed; the usual Gini-Simpson criterion presents difficulties when there is a relatively large number of classes and improved splitting criteria are obtained. One particular improvement is the introduction of adaptive anti-end-cut factors that take advantage of highly asymmetrical splits where appropriate. They use the number and mix of classes in the current node of the tree to identify whether or not it is likely to be advantageous to create a very small offspring node. A number of data sets are used as examples.     

The rhetorical construction of efficiency: Restructuring and industrial democracy in Mondragón,Spain

This paper examines the tension between the struggle to survive in a competitive economy and efforts to create more democratic workplaces in the Basque worker cooperative complex in Mondragón, Spain. Data from archival research, participant observation, and interviews are used to outline the historical context of structural change and policy-making in the complex. Language used in internal policy discussions is analyzed to explore the way in which arguments about efficiency have shaped interpretations of structural change and these organizations' policy responses. It is argued that in these cooperatives, the concept of efficiency does not operate as a neutral, objective benchmark of organizational performance, but is socially and rhetorically constructed. The predominant approach to efficiency that is emerging in Mondragón aims to make the cooperatives more competitive, but also facilitates a gradual displacement of collectively established organizational objectives such as relative equality, job security, and favorable work conditions for ones that privilege institutional interests of profitability and stability. More importantly, it has helped usher in important changes in the policy-making process itself, limiting the range of legitimate participation and weakening broad-based control over the identification of appropriate objectives and means. The paper suggests that reframing an existing rhetorical strategy to highlight more clearly the link between efficiency and any given set of objectives would stimulate a broader discussion of policy issues and help balance more effectively the diverse interests at stake in Mondragón's restructuring.An earlier version of this paper was presented at the 1992 annual meetings of the American Sociological Association in Pittsburgh, Pennsylvania.