Broadband prices in the European Union : competition and commercial strategies

This paper analyses the determinants of broadband Internet access prices in a group of 15 EU countries between 2008 and 2011. Using a rich panel dataset of broadband plans, we show the positive effect of downstream speed on prices, and report that cable and fibre-to-the-home technologies are available at lower prices per Mbps than x DSL technology. Operators’marketing strategies are also analysed as we show how much prices rise when the broadband service is offered in a bundle with voice telephony and/or television, and how much they fall when download volume caps are included. The most insightful results of this study are provided by a group of metrics that represent the situation of competition and entry patterns in the broadband market. We show that consumer segmentation positively affects prices. On the other hand, broadband prices are higher in countries where entrants make greater use of bitstream access and lower when they use more intensively direct access -local loop unbundling-. However, we do not find a significant effect of inter-platform competition on prices.

Automatic Brain Extraction in Fetal MRI using Multi-Atlas-based Segmentation

In fetal brain MRI, most of the high-resolution reconstruction algorithms rely on brain segmentation as a preprocessing step. Manual brain segmentation is however highly time-consuming and therefore not a realistic solution. In this work, we assess on a large dataset the performance of Multiple Atlas Fusion (MAF) strategies to automatically address this problem. Firstly, we show that MAF significantly increase the accuracy of brain segmentation as regards single-atlas strategy. Secondly, we show that MAF compares favorably with the most recent approach (Dice above 0.90). Finally, we show that MAF could in turn provide an enhancement in terms of reconstruction quality.

HouSI: Heuristic for delimitation of housing submarkets and price homogeneous areas

This paper seeks to address the problem of the empirical identification of housing market segmentation,once we assume that submarkets exist. The typical difficulty in identifying housing submarkets when dealing with many locations is the vast number of potential solutions and, in such cases, the use of the Chow test for hedonic functions is not a practical solution. Here, we solve this problem by undertaking an identification process with a heuristic for spatially constrained clustering, the"Housing Submarket Identifier" (HouSI). The solution is applied to the housing market in the city of Barcelona (Spain), where we estimate a hedonic model for fifty thousand dwellings aggregated into ten groups. In order to determine the utility of the procedure we seek to verify whether the final solution provided by the heuristic is comparable with the division of the city into ten administrative districts.

Redox dysregulation in schizophrenia : effect on myelination of cortical structures and connectivity

Cette thèse traite du rôle qu'un facteur de risque génétique développé chez les patients souffrant de schizophrénie, à savoir un déficit de la synthèse du glutathion, peut jouer dans les anomalies de la connectivité cérébrale trouvées chez ces patients. L'essentiel du travail a été consacré à évaluer la structure de la substance blanche dans l'ensemble du cerveau chez un modèle animal par une méthode similaire à celle utilisée en recherche clinique avec l'imagerie par résonance magnétique (IRM). Cette approche de translation inverse chez la souris knock-out de glutamate-cystéine ligase modulateur sous-unité (Gclm KO), avait l'objectif d'étudier l'effet des défenses redox déficientes sur le développement des connexions cérébrales, tout en excluant celui des facteurs non liés au génotype. Après avoir établi le protocole de recherche, l'influence d'une manipulation environnementale a également été étudiée. Pour effectuer une analyse statistique fiable des données d'IRM obtenues, nous .avons d'abord créé un atlas du cerveau de la souris afin de l'utiliser comme modèle pour une segmentation précise des différentes régions du cerveau sur les images IRM obtenues in vivo. Les données provenant de chaque région d'intérêt ont ensuite été étudiées séparément. La qualité de cette méthode a été évaluée dans une expérience de simulation pour déduire la puissance statistique réalisable dans chaque région en fonction du nombre d'animaux utilisés. Ces outils d'analyse nous ont permis d'évaluer l'intégrité de la substance blanche dans le cerveau des souris durant le développement grâce à une expérience longitudinale, en utilisant l'imagerie du tenseur de diffusion (DTI). Nous avons ainsi observé des anomalies dans les paramètres dérivés du tenseur (diffusivité et anisotropie) dans la Commissure Antérieure et le Fimbria/Fornix des souris Gclm KO, par rapport aux animaux contrôles. Ces résultats suggèrent une substance blanche endommagée dans ces régions. Dans une expérience électrophysiologique, Pascal Steullet a montré que ces anomalies ont des conséquences fonctionnelles caractérisées par une réduction de la vitesse de conduction dans les fibres nerveuses. Ces données renforcent les conclusions des analyses d'imagerie. Le mécanisme par lequel une dérégulation redox affecte la structure de la substance blanche reste encore à définir, car une analyse immunohistochimique des protéines constituantes de la couche de myéline des fibres concernées n'a pas donné de résultats concluants. Nous avons également constaté un élargissement des ventricules dans les jeunes souris Gclm KO, mais pas chez les adultes et des anomalies neurochimiques déjà connues chez ces animaux (Duarte et al. 2011), à savoir une réduction du Glutathion et une augmentation de l'acide N-acétylaspartique, de l'Alanine et du ratio Glutamine/Glutamate. Nous avons ensuite testé l'effet d'un stress environnemental supplémentaire, l'élevage en isolement social, sur le phénotype. Ce stress n'a eu aucun effet sur la structure de la substance blanche évaluée par DTI, mais a réduit la concentration de myo-Inositol et augmenté le ratio de Glutamine/Glutamate dans le cortex frontal. Nous avons aussi reproduit dans ce groupe indépendant d'animaux les effets du génotype sur le profil neurochimique, sur la taille des ventricules et aussi sur les paramètres dérivés du tenseur de diffusion dans le Fimbria/Fornix, mais pas dans la Commissure Antérieure. Nos résultats montrent qu'une dérégulation redox d'origine génétique perturbe la structure et la fonction de la substance blanche dans des régions spécifiques, causant ainsi l'élargissement des ventricules. Ces phénotypes rassemblent certaines caractéristiques neuro-anatomiques de la schizophrénie, mais les mécanismes qui en sont responsables demeurent encore inconnus. L'isolement social n'a pas d'effet sur la structure de la substance blanche évaluée par DTI, alors qu'il est prouvé qu'il affecte la maturation des oligodendrocytes. La neurochimie corticale et en particulier le rapport Glutamine/Glutamate a été affecté par le dérèglement redox ainsi que par l'isolement social. En conséquence, ce ratio représente un indice prometteur dans la recherche sur l'interaction du stress environnemental avec le déséquilibre redox dans le domaine de la schizophrénie. -- The present doctoral thesis is concerned with the role that a genetic risk factor for the development of schizophrenia, namely a deficit in Glutathione synthesis, may play in the anomalies of brain connectivity found in patients. Most of the effort was devoted to perform a whole-brain assessment of white matter structure in the Glutamate-Cysteine ligase modulatory knockout mouse model (Gclm KO) using Magnetic Resonance Imaging (MRI) techniques similar to those used in state-of-the-art clinical research. Such reverse translational approach taking brain imaging from the bedside to the bench aimed to investigate the role that deficient redox defenses may play in the development of brain connections while excluding all influencing factors beside the genotype. After establishing the protocol, the influence of further environmental manipulations was also studied. Analysis of MRI images acquired in vivo was one of the main challenges of the project. Our strategy consisted in creating an atlas of the mouse brain to use as segmentation guide and then analyze the data from each region of interest separately. The quality of the method was assessed in a simulation experiment by calculating the statistical power achievable in each brain region at different sample sizes. This analysis tool enabled us to assess white matter integrity in the mouse brain along development in a longitudinal experiment using Diffusion Tensor Imaging (DTI). We discovered anomalies in diffusivity parameters derived from the tensor in the Anterior Commissure and Fimbria/Fornix of Gclm KO mice when compared to wild-type animals, which suggest that the structure of these tracts is compromised in the KO mice. In an elegant electrophysiological experiment, Pascal Steullet has provided evidence that these anomalies have functional consequences in form of reduced conduction velocity in the concerned tracts, thus supporting the DTI findings. The mechanism by which redox dysregulation affects WM structure remains unknown, for the immunohistochemical analysis of myelin constituent proteins in the concerned tracts produced inconclusive results. Our experiments also detected an enlargement of the lateral ventricles in young but not adult Gclm KO mice and confirmed neurochemical anomalies already known to affect this animals (Duarte et al. 2011), namely a reduction in Glutathione and an increase in Glutamine/Glutamate ratio, N-acetylaspartate and Alanine. Using the same methods, we tested the effect of an additional environmental stress on the observed phenotype: rearing in social isolation had no effect on white matter structure as assessed by DTI, but it reduced the concentration of myo-Inositol and increased the Glutamine/Glutamate ratio in the frontal cortex. We could also replicate in this separate group of animals the effects of genotype on the frontal neurochemical profile, ventricular size and diffusivity parameters in the Fimbria/Fornix but not in the Anterior Commissure. Our data show that a redox dysregulation of genetic origin may disrupt white matter structure and function in specific tracts and cause a ventricular enlargement, phenotypes that resemble some neuroanatomical features of schizophrenia. The mechanism responsible remains however unknown. We have also demonstrated that environmental stress in form of social isolation does not affect white matter structure as assessed by DTI even though it is known to affect oligodendrocyte maturation. Cortical neurochemistry, and specifically the Glutamine to Glutamate balance was affected both by redox dysregulation and social isolation, and is thus a good target for further research on the interaction of redox imbalance and environmental stress in schizophrenia.

Unsupervised active regions for multiresolution image segmentation

An unsupervised approach to image segmentation which fuses region and boundary information is presented. The proposed approach takes advantage of the combined use of 3 different strategies: the guidance of seed placement, the control of decision criterion, and the boundary refinement. The new algorithm uses the boundary information to initialize a set of active regions which compete for the pixels in order to segment the whole image. The method is implemented on a multiresolution representation which ensures noise robustness as well as computation efficiency. The accuracy of the segmentation results has been proven through an objective comparative evaluation of the method

Using appearance and context for outdoor scene object classification

We propose a probabilistic object classifier for outdoor scene analysis as a first step in solving the problem of scene context generation. The method begins with a top-down control, which uses the previously learned models (appearance and absolute location) to obtain an initial pixel-level classification. This information provides us the core of objects, which is used to acquire a more accurate object model. Therefore, their growing by specific active regions allows us to obtain an accurate recognition of known regions. Next, a stage of general segmentation provides the segmentation of unknown regions by a bottom-strategy. Finally, the last stage tries to perform a region fusion of known and unknown segmented objects. The result is both a segmentation of the image and a recognition of each segment as a given object class or as an unknown segmented object. Furthermore, experimental results are shown and evaluated to prove the validity of our proposal

Data mining and mall users profile

Marketing scholars have suggested a need for more empirical research on consumer response to malls, in order to have a better understanding of the variables that explain the behavior of the consumers. The segmentation methodology CHAID (Chi-square automatic interaction detection) was used in order to identify the profiles of consumers with regard to their activities at malls, on the basis of socio-demographic variables and behavioral variables (how and with whom they go to the malls). A sample of 790 subjects answered an online questionnaire. The CHAID analysis of the results was used to identify the profiles of consumers with regard to their activities at malls. In the set of variables analyzed the transport used in order to go shopping and the frequency of visits to centers are the main predictors of behavior in malls. The results provide guidelines for the development of effective strategies to attract consumers to malls and retain them there.

Image Segmentation Using Information Bottleneck Method

In image processing, segmentation algorithms constitute one of the main focuses of research. In this paper, new image segmentation algorithms based on a hard version of the information bottleneck method are presented. The objective of this method is to extract a compact representation of a variable, considered the input, with minimal loss of mutual information with respect to another variable, considered the output. First, we introduce a split-and-merge algorithm based on the definition of an information channel between a set of regions (input) of the image and the intensity histogram bins (output). From this channel, the maximization of the mutual information gain is used to optimize the image partitioning. Then, the merging process of the regions obtained in the previous phase is carried out by minimizing the loss of mutual information. From the inversion of the above channel, we also present a new histogram clustering algorithm based on the minimization of the mutual information loss, where now the input variable represents the histogram bins and the output is given by the set of regions obtained from the above split-and-merge algorithm. Finally, we introduce two new clustering algorithms which show how the information bottleneck method can be applied to the registration channel obtained when two multimodal images are correctly aligned. Different experiments on 2-D and 3-D images show the behavior of the proposed algorithms

Analysis of Satellite Data for Forest Segmentation

In this work we study the classification of forest types using mathematics based image analysis on satellite data. We are interested in improving classification of forest segments when a combination of information from two or more different satellites is used. The experimental part is based on real satellite data originating from Canada. This thesis gives summary of the mathematics basics of the image analysis and supervised learning , methods that are used in the classification algorithm. Three data sets and four feature sets were investigated in this thesis. The considered feature sets were 1) histograms (quantiles) 2) variance 3) skewness and 4) kurtosis. Good overall performances were achieved when a combination of ASTERBAND and RADARSAT2 data sets was used.

Country Risk and Market Reactions to Earnings Announcements: Evidence from Finnish Stock Markets

The objective of this thesis is to examine the market reaction around earnings announcements in Finnish stock markets. The aim is to find out whether the extreme market conditions during the financial crisis are reflected in stock prices as a stronger reaction. In addition to this, the purpose is to investigate how extensively Finnish listed companies report the country segmentation of revenues in their interim reports and whether the country risk is having a significant impact on perceived market reaction. The sample covers all companies listed in Helsinki stock exchange at 1.1.2010 and these companies’ interim reports from the first quarter of 2008 to last quarter of 2009. Final sample consists of 81 companies and 630 firm-quarter observations. The data sample has been divided in two parts, of which country risk sample contains 17 companies and 127 observations and comparison sample covers 66 companies and 503 observations. Research methodologies applied in this thesis are event study and cross-sectional regression analysis. Empirical results indicate that the market reaction occurs mainly during the announcement day and is slightly stronger in case of positive earnings surprises than the reactions observed in previous studies. In case of negative earnings surprises no significant differences can be observed. In case of country risk sample and negative earnings surprise market reaction is negative already in advance of the disclosure contrary to comparison sample. In case of positive surprise no differences can be observed. Country risk variable developed during this study seems to explain only minor part of the market reaction.

Segmentation of Industrial Customers – Case: Finnish Mechanical Engineering Companies

Segmentointi on perinteisesti ollut erityisesti kuluttajamarkkinoinnin työkalu, mutta siirtymä tuotteista palveluihin on lisännyt segmentointitarvetta myös teollisilla markkinoilla. Tämän tutkimuksen tavoite on löytää selkeästi toisistaan erottuvia asiakasryhmiä suomalaisen liikkeenjohdon konsultointiyritys Synocus Groupin tarjoaman case-materiaalin pohjalta. K-means-klusteroinnin avulla löydetään kolme potentiaalista markkinasegmenttiä perustuen siihen, mitkä tarjoamaelementit 105 valikoitua suomalaisen kone- ja metallituoteteollisuuden asiakasta ovat maininneet tärkeimmiksi. Ensimmäinen klusteri on hintatietoiset asiakkaat, jotka laskevat yksikkökohtaisia hintoja. Toinen klusteri koostuu huolto-orientoituneista asiakkaista, jotka laskevat tuntikustannuksia ja maksimoivat konekannan käyttötunteja. Tälle kohderyhmälle kannattaisi ehkä markkinoida teknisiä palveluja ja huoltosopimuksia. Kolmas klusteri on tuottavuussuuntautuneet asiakkaat, jotka ovat kiinnostuneita suorituskyvyn kehittämisestä ja laskevat tonnikohtaisia kustannuksia. He tavoittelevat alempia kokonaiskustannuksia lisääntyneen suorituskyvyn, pidemmän käyttöiän ja alempien huoltokustannusten kautta.

Characterization of fiber and vessel elements in pulp suspension images

The thesis is related to the topic of image-based characterization of fibers in pulp suspension during the papermaking process. Papermaking industry is focusing on process control optimization and automatization, which makes it possible to manufacture highquality products in a resource-efficient way. Being a part of the process control, pulp suspension analysis allows to predict and modify properties of the end product. This work is a part of the tree species identification task and focuses on analysis of fiber parameters in the pulp suspension at the wet stage of paper production. The existing machine vision methods for pulp characterization were investigated, and a method exploiting direction sensitive filtering, non-maximum suppression, hysteresis thresholding, tensor voting, and curve extraction from tensor maps was developed. Application of the method to the microscopic grayscale pulp images made it possible to detect curves corresponding to fibers in the pulp image and to compute their morphological characteristics. Performance of the method was evaluated based on the manually produced ground truth data. An accuracy of fiber characteristics estimation, including length, width, and curvature, for the acacia pulp images was found to be 84, 85, and 60% correspondingly.

Customer Segmentation in Industrial Markets - CASE: Labelstock Company

The aim of this thesis is to study segmentation in industrial markets and develop a segmenting method proposal and criteria case study for a labelstock manufacturing company. An industrial company is facing many different customers with varying needs. Market segmentation is a process for dividing a market into smaller groups in which customers have the same or similar needs. Segmentation gives tools to the marketer to better match the product or service more closely to the needs of the target market. In this thesis a segmentation tool proposal and segmenting criteria is case studied for labelstock company’s Europe, Middle East and Africa business area customers and market. In the developed matrix tool different customers are planned to be evaluated based on customer characteristic variables. The criteria for the evaluating matrix are based on the customer’s buying organizations characteristics and buying behaviour. There are altogether 13 variables in the evaluating matrix. As an example of variables there are loyalty, size of the customer, estimated growth of the customer purchases and customer’s decision-making and buying behaviour. These characteristic variables will help to identify market segments to target and the customers belonging to those segments.

Supplier Segmentation for Nordic Petrochemical Industry

Objective of this thesis was to map possibilities for systematic supplier management in field of chemical process industry. Through this study it was aimed to develop a tool for supplier management that could be integrated with operations in business unit. With developed tool suppliers should be able to be segmented based on their willingness and capability, and segmentation could be applied in purchasing decisions. In this thesis there was made a survey of methods that are recognized in literature to manage and allocate suppliers. This thesis recognizes segmentation as a method to group and select suppliers in procurement. Based on literature, a proposal for segmentation framework and evaluation criteria factors will be constituted. Based on theoretical proposal, in an expertise workshop a final segmentation framework was constituted, which covers segments with descriptions and evaluation part. Evaluation part includes an evaluation framework which helps to score suppliers with selected factors and leads to total grades in willingness and capability. These total grades will be the coordinates and they determine the segment where the supplier under evaluation belongs. In this thesis segments definitions, objectives, and road maps will be described.

Target audience and message alignment for integrated marketing communication of luxury products: entering the Chinese market

The purpose of this study is to identify opportunities to match marketing communication message strategies with the target audience characteristics in the Chinese luxury market entry context. Therefore, consumer behaviour and psychographic marketing segmentation fields are being reviewed in a holistic view in order to identify the similarities and connection points. Through the analysis of the messages in advertisements placed in a certain luxury and fine living magazine, message creation strategies are being anticipated.