1,721,060 research outputs found
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Modèles probabilistes pour la prévision de l'offre et de la demande dans le secteur de l'énergie électrique
No full textCette thèse étudie des modèles prédictifs probabilistes basés sur les processus gaussiens et l'apprentissage en profondeur pour la prévision de la demande d'électricité. Étant donné que les processus gaussiens sont des modèles prédictifs basés sur des noyaux, leur performance est limitée par le type, le nombre et la dimension du noyau sélectionné. Pour répondre à ces limitations, premièrement, elle propose une nouvelle technique d'approximation gaussienne qui aborde le goulot d'étranglement computationnel bayésien. Deuxièmement, elle propose un algorithme d'estimation de noyau compositionnel stochastique en utilisant l'approximation gaussienne proposée comme modèle sous-jacent. Troisièmement, elle suit une procédure itérative utilisant la validation croisée pour sélectionner les meilleures combinaisons de noyaux qui expliquent le mieux le modèle de génération des données. De plus, elle tente également de pallier la limitation de l'approche du maximum de vraisemblance, qui est généralement utilisée dans les modèles probabilistes d'apprentissage en profondeur et qui ne garantit pas nécessairement une largeur d'intervalle minimisée et une probabilité de couverture maximisée pour les points prévus. Cette thèse propose un nouvel algorithme d'entraînement pour les réseaux neuronaux. L'algorithme proposé d'estimation des bornes inférieures et supérieures basé sur la distribution englobe la largeur d'intervalle et la probabilité de couverture en tant que mesures de qualité avec des paramètres adaptatifs garantissant les performances nécessaires par rapport à d'autres techniques alternatives. Les approches suggérées renforcent le déploiement des modèles gaussiens et d'apprentissage en profondeur dans le secteur de l'énergie. Le modèle d'estimation des bornes pour un intervalle de prédiction minimisé et une probabilité de couverture maximisée peut aider les fournisseurs d'énergie potentiels à dimensionner les générateurs, ce qui se traduira par un gain de profit marginal. De plus, l'algorithme d'estimation de noyau peut simplifier l'application de l'apprentissage basé sur les noyaux pour ceux qui trouvent la sélection des noyaux vague. Pour les experts, il peut donner un aperçu préliminaire de la structure des noyaux qui pourraient potentiellement s'adapter aux données. La technique d'échantillonnage aléatoire de colonnes pourrait offrir une méthode alternative pour la construction et l'approximation rapide d'un modèle gaussien, scalable à de grandes données. De plus, l'estimation des bornes, en plus de fournir une distribution de prévision aux modèles neuronaux à estimation de point, peut également servir de point de départ pour une formation de modèle probabiliste alternative dans les réseaux neuronaux profondsThis thesis investigate probabilistic predictive models based on the Gaussian process and deep learning for electricity demand forecasting. As Gaussian processes are kernel-based predictive models, their performance is constrained by the type, number and dimension of the selected kernel. To address these limitations, first it proposes a new Gaussian approximation technique that address the Bayesian computational bottleneck. Second, it proposes a stochastic compositional kernel estimation algorithm using the proposed Gaussian approximation as the underlying model. Third, it follows an iterative procedure using cross-validation for selecting an optimal combinations of kernels that best explain the data generating model. Furthermore, it also tries to address the limitation of maximum likelihood approach which is usually employed in probabilistic deep learning models and yet fails in guaranteeing a minimized interval width and maximized coverage probability for the forecasted points. This thesis proposes a new training algorithm for neural networks. The proposed distribution based lower upper bound estimation algorithm encompasses interval width and coverage probability as quality metrics with adaptive parameters that guarantee the needed performance compared to other alternative techniques. The suggested approaches enhance the deployment of Gaussian and deep learning models in the energy sector. The bound estimation model for a minimized prediction interval and maximized coverage probability, can help potential energy suppliers in sizing generators which will result in a marginal profit gain. In addition, the kernel estimation algorithm can simplify the application of kernel-based learning to those who find kernel selection vague. To the experienced, it can give a preliminary insight into the structure of the kernels that could potentially fit the data. The randomized column sampling technique could offer an alternative method for a fast Gaussian model building and approximation that is scalable to large data. Furthermore, the bound estimation, in addition to providing a forecast distribution to a point estimate neural models, it can also serve as a good starting point to an alternative probabilistic model training in deep neural net
Routing Bandwidth Guaranteed paths for QoS Flows in Ad Hoc Networks under Interferences Influence
No full textInternational audienc
A Quasipolynomial-Time and deterministic Source-Based Heuristic for Multicasting Multimedia Information
No full textInternational audienc
QMOST: A QoSaware Multicast Overlay Spanning Tree Proto-col for Multimedia Applications in Mobile Ad hoc Networks
No full textInternational audienc
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
