La creciente importancia de las redes sociales online para la compra-venta de productos y servicios.

En este proyecto se plantea un modelo que relacione directamente la popularidad o actividad de las empresas en redes sociales con su facturación. Español

On the optimal usage of labelled examples in semi-supervised multi-class classification problems

In recent years, the performance of semi-supervised learning has been theoretically investigated. However, most of this theoretical development has focussed on binary classification problems. In this paper, we take it a step further by extending the work of Castelli and Cover [1] [2] to the multi-class paradigm. Particularly, we consider the key problem in semi-supervised learning of classifying an unseen instance x into one of K different classes, using a training dataset sampled from a mixture density distribution and composed of l labelled records and u unlabelled examples. Even under the assumption of identifiability of the mixture and having infinite unlabelled examples, labelled records are needed to determine the K decision regions. Therefore, in this paper, we first investigate the minimum number of labelled examples needed to accomplish that task. Then, we propose an optimal multi-class learning algorithm which is a generalisation of the optimal procedure proposed in the literature for binary problems. Finally, we make use of this generalisation to study the probability of error when the binary class constraint is relaxed.

Antecedentes de la institucionalización de las organizaciones

[ES] Esta investigación es un intento de avanzar en la comprensión de por qué las organizaciones son sensibles a la institucionalización. Para ello, describimos los elementos clave que ayudan a explicar el origen del proceso corporativo de institucionalización. Además, se ha seguido una metodología de investigación cualitativa, utilizando la técnica del concept mapping , para agrupar en constructos los diferentes ítems que actúan como factores motivadores de la transformación de las organizaciones en instituciones. Metodológicamente hemos tratado de obviar la separación entre viejo y nuevo institucionalismo siguiendo a los autores que cuestionan la conveniencia de trazar una línea divisoria entre la vieja y la nueva teoría. Consideramos que el papel del CEO es esencial en el impulso del proceso de institucionalización, aunque en muchas ocasiones sus decisiones estén apoyadas o hayan pasado por el filtro de los equipos de gobierno de la organización o de los consejos de administración. Cualquier impulso que realice la organización dependerá fundamentalmente de las capacidades, las sensaciones, la formación y el modo de pensar del CEO. Los resultados refuerzan varios de los temas claves sugeridos en la literatura sobre Teoría Institucional. En particular, se establece una clasificación con los motivos que dan origen a las iniciativas institucionales, a saber: autoridad institucional; ventaja en gestión; e, implicación social. Esta clasificación es coincidente, en una gran medida, con los pilares de la institucionalización que han sido definidos en la literatura de la teoría institucional, ayudando a comprender, con mayor detalle, el origen de los procesos institucionales y los antecedentes o motivaciones que los generan y guían.

Online journalism: research methods. A multidisciplinary approach in comparative perspective

Journalism on digital networks, and specifically on Internet, is a relatively recent phenomenon, whose spread began in approximately 1994, in parallel to that of the World Wide Web. Study of this new communicative phenomenon began simultaneously in several countries. This, in its turn, was helped by the new possibilities for communication amongst academics - electronic mail, predating the WWW, was, and is, one of the most widely used tools of the university community. The spread of these new forms of global communication helped to raise mutual awareness between research groups, making it possible to form increasingly broad and cohesive networks.

Construcción de un modelo predictivo de tipo Gradient boosting para ventas online

En este proyecto se describirá como construir un modelo predictivo de tipo gradient boosting para predecir el número de ventas online de un producto X del cual solo sabremos su número de identificación, teniendo en cuenta las campañas publicitarias y las características tanto cualitativas y cuantitativas de éste. Para ello se utilizarán y se explicarán las diferentes técnicas utilizadas, como son: la técnica de la validación cruzada y el Blending. El objetivo del proyecto es implementar el modelo así como explicar con exactitud cada técnica y herramienta utilizada y obtener un resultado válido para la competición propuesta en Kaggle con el nombre de Online Product Sales.

Integrated treatment of first episode psychosis with online training (e-learning): study protocol for a randomised controlled trial

Background: The integrated treatment of first episode psychosis has been shown to improve functionality and negative symptoms in previous studies. In this paper, we describe a study of integrated treatment (individual psychoeducation complementary to pharmacotherapy) versus treatment as usual, comparing results at baseline with those at 6-month re-assessment (at the end of the study) for these patients, and online training of professionals to provide this complementary treatment, with the following objectives: 1) to compare the efficacy of individual psychoeducation as add-on treatment versus treatment as usual in improving psychotic and mood symptoms; 2) to compare adherence to medication, functioning, insight, social response, quality of life, and brain-derived neurotrophic factor, between both groups; and 3) to analyse the efficacy of online training of psychotherapists. Methods/design: This is a single-blind randomised clinical trial including patients with first episode psychosis from hospitals across Spain, randomly assigned to either a control group with pharmacotherapy and regular sessions with their psychiatrist (treatment as usual) or an intervention group with integrated care including treatment as usual plus a psychoeducational intervention (14 sessions). Training for professionals involved at each participating centre was provided by the coordinating centre (University Hospital of Alava) through video conferences. Patients are evaluated with an extensive battery of tests assessing clinical and sociodemographic characteristics (Positive and Negative Syndrome Scale, State-Trait Anxiety Inventory, Liebowitz Social Anxiety Scale, Hamilton Rating Scale for Depression, Scale to Assess Unawareness of Mental Disorders, Strauss and Carpenter Prognostic Scale, Global Assessment of Functioning Scale, Morisky Green Adherence Scale, Functioning Assessment Short Test, World Health Organization Quality of Life instrument WHOQOL-BREF (an abbreviated version of the WHOQOL-100), and EuroQoL questionnaire), and brain-derived neurotrophic factor levels are measured in peripheral blood at baseline and at 6 months. The statistical analysis, including bivariate analysis, linear and logistic regression models, will be performed using SPSS. Discussion: This is an innovative study that includes the assessment of an integrated intervention for patients with first episode psychosis provided by professionals who are trained online, potentially making it possible to offer the intervention to more patients.

Contributions on distance-based algorithms, multi-classifier construction and pairwise classification

179 p.

Contributions to learning Bayesian network models from weakly supervised data: Application to Assisted Reproductive Technologies and Software Defect Classification

162 p.

El discurso en el estudio de las organizaciones. Una revisión de la literatura

[ES] En el documento se presenta una revisión de la literatura referente al discurso en el estudio de las organizaciones, que parte de sus fundamentos teóricos y sus principales aportes. La unidad de análisis fueron 88 textos seleccionados en una revisión sistemática de 5.509 artículos, publicados entre 2000 y 2012, en diez de las principales revistas internacionales relacionadas con la temática definida. Para relevar y categorizar los datos pertinentes, se utilizó el software de investigación cualitativa Atlas.ti, versión 6.2. La comprensión de la información se complementó con métodos de análisis de contenido y de frecuencia. De los principales resultados se infirieron cuatro dimensiones del discurso: instrumental, simbólica, estratégica y comunicacional. Además, se realizó una clasificación de los artículos con base en los enfoques sobre el discurso y las temáticas organizacionales estudiadas. Los aportes encontrados en dicha revisión permiten plantear que el discurso no solo es lo dicho sino también aquello que ha sido silenciado en la organización, campo de investigación que ha sido poco explorado en los textos analizados. Este artículo realiza contribuciones importantes para el área administrativa y organizacional, en especial para todos aquellos profesionales que se interesen por el análisis de las organizaciones desde la perspectiva del discurso, tanto en la academia como en la asesoría e intervención en ellas.

Image-Based Delineation and Classification of Built Heritage Masonry

Fundacion Zain is developing new built heritage assessment protocols. The goal is to objectivize and standardize the analysis and decision process that leads to determining the degree of protection of built heritage in the Basque Country. The ultimate step in this objectivization and standardization effort will be the development of an information and communication technology (ICT) tool for the assessment of built heritage. This paper presents the ground work carried out to make this tool possible: the automatic, image-based delineation of stone masonry. This is a necessary first step in the development of the tool, as the built heritage that will be assessed consists of stone masonry construction, and many of the features analyzed can be characterized according to the geometry and arrangement of the stones. Much of the assessment is carried out through visual inspection. Thus, this process will be automated by applying image processing on digital images of the elements under inspection. The principal contribution of this paper is the automatic delineation the framework proposed. The other contribution is the performance evaluation of this delineation as the input to a classifier for a geometrically characterized feature of a built heritage object. The element chosen to perform this evaluation is the stone arrangement of masonry walls. The validity of the proposed framework is assessed on real images of masonry walls.

Towards Application of One-Class Classification Methods to Medical Data

In the problem of one-class classification (OCC) one of the classes, the target class, has to be distinguished from all other possible objects, considered as nontargets. In many biomedical problems this situation arises, for example, in diagnosis, image based tumor recognition or analysis of electrocardiogram data. In this paper an approach to OCC based on a typicality test is experimentally compared with reference state-of-the-art OCC techniques-Gaussian, mixture of Gaussians, naive Parzen, Parzen, and support vector data description-using biomedical data sets. We evaluate the ability of the procedures using twelve experimental data sets with not necessarily continuous data. As there are few benchmark data sets for one-class classification, all data sets considered in the evaluation have multiple classes. Each class in turn is considered as the target class and the units in the other classes are considered as new units to be classified. The results of the comparison show the good performance of the typicality approach, which is available for high dimensional data; it is worth mentioning that it can be used for any kind of data (continuous, discrete, or nominal), whereas state-of-the-art approaches application is not straightforward when nominal variables are present.