Determinants of entrepreneurship: Are women different?

In this paper we investigate, for the first time, how individual determinants of entrepreneurship - such as age, income, education, work status, skills, access to networks and fear of failure - differ between males and females. We conduct our exercise using individual data provided by the Global Entrepreneurship Monitor (GEM), available for 46 countries, between 2001 and 2004. The literature on entrepreneurship has uncovered differences in the rate of entrepreneurship between men and women, with women generally displaying lower entrepreneurial activity than men. This is important since, as we show, entrepreneurial activity is positively related across countries with the female to male entrepreneurial ratio. We examine total entrepreneurship rates, as well as entrepreneurship driven by opportunity and by need. We find that indeed entrepreneurial activity rates are lower for females across all but one of the countries in the sample. Looking at categorical groups – by age interval, education, work status, etc. – we find that female entrepreneurial rates are significantly lower than for males. For the first time we test for differences in the characteristics of female and male entrepreneurs and find that female entrepreneurs are slightly older, more frequently at home or not working, lower income and lower educated, and less access to business networks than their male counterparts. AS to the determinants of entrepreneurial rates themselves, the main differences across genders are the lower impact of secondary education and the larger impact of skills and fear of failure in female entrepreneurial rates relative to males. Results for entrepreneurship by opportunity and by necessity confirm the larger importance of specific skills for women creating new businesses,. Our results suggest that facilitating access to business networks and specific business skills are the most powerful instruments to increase the rates of female entrepreneurship.

Modeling High-Dimensional Audio Sequences with Recurrent Neural Networks

Cette thèse étudie des modèles de séquences de haute dimension basés sur des réseaux de neurones récurrents (RNN) et leur application à la musique et à la parole. Bien qu'en principe les RNN puissent représenter les dépendances à long terme et la dynamique temporelle complexe propres aux séquences d'intérêt comme la vidéo, l'audio et la langue naturelle, ceux-ci n'ont pas été utilisés à leur plein potentiel depuis leur introduction par Rumelhart et al. (1986a) en raison de la difficulté de les entraîner efficacement par descente de gradient. Récemment, l'application fructueuse de l'optimisation Hessian-free et d'autres techniques d'entraînement avancées ont entraîné la recrudescence de leur utilisation dans plusieurs systèmes de l'état de l'art. Le travail de cette thèse prend part à ce développement. L'idée centrale consiste à exploiter la flexibilité des RNN pour apprendre une description probabiliste de séquences de symboles, c'est-à-dire une information de haut niveau associée aux signaux observés, qui en retour pourra servir d'à priori pour améliorer la précision de la recherche d'information. Par exemple, en modélisant l'évolution de groupes de notes dans la musique polyphonique, d'accords dans une progression harmonique, de phonèmes dans un énoncé oral ou encore de sources individuelles dans un mélange audio, nous pouvons améliorer significativement les méthodes de transcription polyphonique, de reconnaissance d'accords, de reconnaissance de la parole et de séparation de sources audio respectivement. L'application pratique de nos modèles à ces tâches est détaillée dans les quatre derniers articles présentés dans cette thèse. Dans le premier article, nous remplaçons la couche de sortie d'un RNN par des machines de Boltzmann restreintes conditionnelles pour décrire des distributions de sortie multimodales beaucoup plus riches. Dans le deuxième article, nous évaluons et proposons des méthodes avancées pour entraîner les RNN. Dans les quatre derniers articles, nous examinons différentes façons de combiner nos modèles symboliques à des réseaux profonds et à la factorisation matricielle non-négative, notamment par des produits d'experts, des architectures entrée/sortie et des cadres génératifs généralisant les modèles de Markov cachés. Nous proposons et analysons également des méthodes d'inférence efficaces pour ces modèles, telles la recherche vorace chronologique, la recherche en faisceau à haute dimension, la recherche en faisceau élagué et la descente de gradient. Finalement, nous abordons les questions de l'étiquette biaisée, du maître imposant, du lissage temporel, de la régularisation et du pré-entraînement.

Estructura del servidor interactivo de información epidemiológica

Los sistemas de software SIG Open Source implementados en servidores Web se encuentran en una etapa de madurez, de forma que son en muchos aspectos iguales o superiores a sus homónimos comerciales; su sofisticación, solidez, estabilidad y manejabilidad facilitan la creación de sistemas de información geográfica accesibles mediante Internet, especializados en diferentes materias, entre ellas la investigación médica. La vigilancia de la situación en salud sigue siendo la piedra angular de la detección de amenazas para la población, que requieren medidas de salud pública. La distribución de este tipo de información especializada es muy útil tanto para apoyar posteriores estudios técnicos, como para cubrir las necesidades de información de los potenciales usuarios. (...)

Semi-automatic knowledge extraction, representation and context-sensitive intelligent retrieval of video content using collateral context modelling with scalable ontological networks

Automatic indexing and retrieval of digital data poses major challenges. The main problem arises from the ever increasing mass of digital media and the lack of efficient methods for indexing and retrieval of such data based on the semantic content rather than keywords. To enable intelligent web interactions, or even web filtering, we need to be capable of interpreting the information base in an intelligent manner. For a number of years research has been ongoing in the field of ontological engineering with the aim of using ontologies to add such (meta) knowledge to information. In this paper, we describe the architecture of a system (Dynamic REtrieval Analysis and semantic metadata Management (DREAM)) designed to automatically and intelligently index huge repositories of special effects video clips, based on their semantic content, using a network of scalable ontologies to enable intelligent retrieval. The DREAM Demonstrator has been evaluated as deployed in the film post-production phase to support the process of storage, indexing and retrieval of large data sets of special effects video clips as an exemplar application domain. This paper provides its performance and usability results and highlights the scope for future enhancements of the DREAM architecture which has proven successful in its first and possibly most challenging proving ground, namely film production, where it is already in routine use within our test bed Partners' creative processes. (C) 2009 Published by Elsevier B.V.

NDVI With Artificial Neural Networks For SRTM Elevation Model Improvement – Hydrological Model Application

Digital elevation model (DEM) plays a substantial role in hydrological study, from understanding the catchment characteristics, setting up a hydrological model to mapping the flood risk in the region. Depending on the nature of study and its objectives, high resolution and reliable DEM is often desired to set up a sound hydrological model. However, such source of good DEM is not always available and it is generally high-priced. Obtained through radar based remote sensing, Shuttle Radar Topography Mission (SRTM) is a publicly available DEM with resolution of 92m outside US. It is a great source of DEM where no surveyed DEM is available. However, apart from the coarse resolution, SRTM suffers from inaccuracy especially on area with dense vegetation coverage due to the limitation of radar signals not penetrating through canopy. This will lead to the improper setup of the model as well as the erroneous mapping of flood risk. This paper attempts on improving SRTM dataset, using Normalised Difference Vegetation Index (NDVI), derived from Visible Red and Near Infra-Red band obtained from Landsat with resolution of 30m, and Artificial Neural Networks (ANN). The assessment of the improvement and the applicability of this method in hydrology would be highlighted and discussed.

Combinando sistemas de informação geográfica e redes neurais GMDH para prever densidades demográficas

The aim of this work is to advance a new approach for estimating demographic density, through combining a Geographic Information System with GMDH Neural Networks. The model that is suggested parts the analyzed space into a rectangular grid formed by multiple cells measuring 0.01 km2 each. The forecasts are elaborated based on the demographic density in each cell and in its neighboring cells at a given time. Despite the limited availability of data during the modeling phase, the utilization of this method for studying a Brazilian medium-sized city presented promising results.

Photovoltaic power forecast modeling with artificial neural networks

Dissertação de Mestrado, Engenharia Eletrónica e Telecomunicações, Faculdade de Ciências e Tecnologia, Universidade do Algarve, 2016

Adaptive Sampling with Mobile Sensor Networks

Mobile sensor networks have unique advantages compared with wireless sensor networks. The mobility enables mobile sensors to flexibly reconfigure themselves to meet sensing requirements. In this dissertation, an adaptive sampling method for mobile sensor networks is presented. Based on the consideration of sensing resource constraints, computing abilities, and onboard energy limitations, the adaptive sampling method follows a down sampling scheme, which could reduce the total number of measurements, and lower sampling cost. Compressive sensing is a recently developed down sampling method, using a small number of randomly distributed measurements for signal reconstruction. However, original signals cannot be reconstructed using condensed measurements, as addressed by Shannon Sampling Theory. Measurements have to be processed under a sparse domain, and convex optimization methods should be applied to reconstruct original signals. Restricted isometry property would guarantee signals can be recovered with little information loss. While compressive sensing could effectively lower sampling cost, signal reconstruction is still a great research challenge. Compressive sensing always collects random measurements, whose information amount cannot be determined in prior. If each measurement is optimized as the most informative measurement, the reconstruction performance can perform much better. Based on the above consideration, this dissertation is focusing on an adaptive sampling approach, which could find the most informative measurements in unknown environments and reconstruct original signals. With mobile sensors, measurements are collect sequentially, giving the chance to uniquely optimize each of them. When mobile sensors are about to collect a new measurement from the surrounding environments, existing information is shared among networked sensors so that each sensor would have a global view of the entire environment. Shared information is analyzed under Haar Wavelet domain, under which most nature signals appear sparse, to infer a model of the environments. The most informative measurements can be determined by optimizing model parameters. As a result, all the measurements collected by the mobile sensor network are the most informative measurements given existing information, and a perfect reconstruction would be expected. To present the adaptive sampling method, a series of research issues will be addressed, including measurement evaluation and collection, mobile network establishment, data fusion, sensor motion, signal reconstruction, etc. Two dimensional scalar field will be reconstructed using the method proposed. Both single mobile sensors and mobile sensor networks will be deployed in the environment, and reconstruction performance of both will be compared.In addition, a particular mobile sensor, a quadrotor UAV is developed, so that the adaptive sampling method can be used in three dimensional scenarios.

Optimization, high availability and redundancy in information communications and technology using networks and systems virtualization architectures

Mestrado em Engenharia Informática

Predicting market movement with Artificial Neural Networks

This paper presents an application of an Artificial Neural Network (ANN) to the prediction of stock market direction in the US. Using a multilayer perceptron neural network and a backpropagation algorithm for the training process, the model aims at learning the hidden patterns in the daily movement of the S&P500 to correctly identify if the market will be in a Trend Following or Mean Reversion behavior. The ANN is able to produce a successful investment strategy which outperforms the buy and hold strategy, but presents instability in its overall results which compromises its practical application in real life investment decisions.

Forecasting of wind speeds and directions with artificial neural networks

In this master’s thesis, wind speeds and directions were modeled with the aim of developing suitable models for hourly, daily, weekly and monthly forecasting. Artificial Neural Networks implemented in MATLAB software were used to perform the forecasts. Three main types of artificial neural network were built, namely: Feed forward neural networks, Jordan Elman neural networks and Cascade forward neural networks. Four sub models of each of these neural networks were also built, corresponding to the four forecast horizons, for both wind speeds and directions. A single neural network topology was used for each of the forecast horizons, regardless of the model type. All the models were then trained with real data of wind speeds and directions collected over a period of two years in the municipal region of Puumala in Finland. Only 70% of the data was used for training, validation and testing of the models, while the second last 15% of the data was presented to the trained models for verification. The model outputs were then compared to the last 15% of the original data, by measuring the mean square errors and sum square errors between them. Based on the results, the feed forward networks returned the lowest generalization errors for hourly, weekly and monthly forecasts of wind speeds; Jordan Elman networks returned the lowest errors when used for forecasting of daily wind speeds. Cascade forward networks gave the lowest errors when used for forecasting daily, weekly and monthly wind directions; Jordan Elman networks returned the lowest errors when used for hourly forecasting. The errors were relatively low during training of the models, but shot up upon simulation with new inputs. In addition, a combination of hyperbolic tangent transfer functions for both hidden and output layers returned better results compared to other combinations of transfer functions. In general, wind speeds were more predictable as compared to wind directions, opening up opportunities for further research into building better models for wind direction forecasting.