1,142 research outputs found
Up quark masses from down quark masses
No full textThe quark and charged lepton masses and the angles and phase of the CKM mixing matrix are nicely reproduced in a model which assumes SU(3)xSU(3) flavour symmetry broken by the v.e.v.'s of fields in its bi-fundamental representation. The relations among the quark mass eigenvalues, m_u/m_c \approx m_c/m_t \approx m^2_d/m^2_s \approx m^2_s/m^2_b \approx \Lambda^2_{GUT}/M^2_{Pl}, follow from the broken flavour symmetry. Large tan(beta) is required which also provides the best fits to data for the obtained textures. Lepton-quark grandunification with a field that breaks both SU(5) and the flavour group correctly extends the predictions to the charged lepton masses. The seesaw extension of the model to the neutrino sector predicts a Majorana mass matrix quadratically hierarchical as compared to the neutrino Dirac mass matrix, naturally yielding large mixings and low mass hierarchy for neutrinos
INTERMEDIATE SYMMETRIES IN THE SPONTANEOUS BREAKING OF SUPERSYMMETRIC SO(10)
No full text We study the supersymmetric spontaneous symmetry breaking of SO(10) into SU (3) ⊗ SU (2) ⊗ U (1) for the most physically interesting cases of SU(5) or flipped SU (5) ⊗ U (1) intermediate symmetries. The first case is more easily realized while the second one requires a fine-tuning condition on the parameters of the superpotential. This is because in the case of SU(5) symmetry there is at most one singlet of the residual symmetry in each SO(10) irreducible representation. We also point out on more general grounds in supersymmetric GUTs that some intermediate symmetries can be exactly realized and others can only be approximated by fine-tuning. In the first category, there could occur some tunneling between the vacua with exact and approximate intermediate symmetry. The flipped SU (5) ⊗ U (1) symmetry improves the unification of gauge couplings if (B-L) is broken by ‖(B-L)‖ =1 scalars yielding right-handed neutrino masses below 1014 GeV . </jats:p
Automatic Author Profiling and Verification
No full textDissertation commitee :
Prof. Jacques Savoy, directeur de thèse, Université de Neuchâtel, Suisse
Dr Valerio Schiavoni, Université de Neuchâtel, Suisse
Prof. Elöd Egyed-Zsigmond, LIRIS, INSA, Lyon, France
Defense : 24th November 2022
No de thèse : 3027Cette thèse s’intéresse principalement aux problèmes de classification de textes fondée sur le style
dont le but est d’identifier les caractéristiques de l’auteur comme son âge, sexe, son idiolecte, en se
basant sur un ensemble de ses écrits. De plus, on aborde la question de savoir si deux textes (comme
des chats, courriels menaçants, testaments douteux, essais, mémos, ou fictions) ont été écrits par la
même personne en comparant leur style d’écriture selon différentes représentations. Nous proposons
un processus de sélection des attributs simple et stable. Cette réduction nous conduit à proposer une
décision possédant un plus grand pouvoir explicatif.
Nous débutons ce travail par analyser l’efficacité de plusieurs modèles basés sur l’apprentissage automatique
et recourant à l’ensemble du vocabulaire. Une procédure de réduction des attributs en deux
étapes peut alors être appliquée. Nous pouvons alors comparer les performances de divers modèles
avec des réduction du nombre d’attributs basés sur notre approche, le 2 ou le PMI. Dans tous les cas,
le nombre d’attributs est réduit à 300.
Sur la base de la collection de documents de différentes campagnes d’évaluation CLEF-PAN, nous
avons testé notre approche avec plusieurs baselines. On constate que les modèles Extra Trees, Random
Forest, ou Gradient Boost produisent souvent les meilleures performances. De plus, la réduction
des attributs au nombre de 300 permet d’obtenir des performances similaires. Cette diminution permet
également de réduire la taille des représentations des documents et donc de réduire le temps de
calcul. Parfois, nous observons même un gain d’efficience.
Dans le cadre de la vérification d’auteur, et selon diverses représentations des textes, nous pouvons
également améliorer la qualité des résultats. Ainsi, les documents présentant de grandes différences
de représentation ne sont pas écrits par la même personne. Dans ce but nous avons appliqué différentes
mesures de performance (AUC, c@1, Final Score (FS)) dont les résultats sont corrélés en
particulier AUC et FS. En tenant compte uniquement du taux de réussite, la pondération TFIDF offre
les meilleures performances.
This thesis mostly discusses the style-based text categorization problem, where the objective is to
identify the author’s demographics, such as gender, age range, and language variety, based on a set of
texts. Also to determine whether two writings (chat, threatening e-mail, doubtful testimony, essays,
text messages, business memos, fanfiction texts) were authored by the same person by contrasting the
writing styles of the two texts by applying the vector difference text representation. We also create a
stable and straightforward paradigm for feature reduction iteratively. This reduction will result to a
more explainable decision.
We begin by assessing the effectiveness of several machine learning models using the complete vocabulary.
The two-step feature selection technique is then used to design a feature reduction strategy.
After testing the models with these reduced features, we were able to examine how the performance
variation would appear in the two scenarios. We went on to test further feature reduction by applying
2 and PMI scoring functions to select the top 300 features.
With the use of several CLEF-PAN datasets, we test our models, and we can see that Extra Trees,
Random Forest, or Gradient Boost often produce the best results. Furthermore, empirical evidence
reveals that the feature set can be effectively condensed using 2 and PMI scoring methods to about
300 features without compromising performance. Additionally, we can see that by discarding non
informative features, decreasing the text feature representation not only cuts down on runtime but
also improves performance in some cases.
With the difference vector text representation approach we demonstrate how utilization of confidencebased
approaches can benefit classification accuracy in the author verification. We can see that small
differences in vectorial representation indicates higher similarity, but documents with a large differences
are not authored by the same writer. Several performance measures are obtained including
accuracy, area under the curve (AUC), c@1 and Final Score (FS). Our research shows a strong correlation
between all performance with measurements FS and AUC having the strongest correlation. We
take into account only the accuracy to draw conclusion about the different text representation methods.
Our experiments therefore show that the best scoring model include TFIDF feature set since it
considers both occurrence frequency and the distribution of terms across the collection
ANIMATE. WP1. T1-1. Primary Sources Savoy Spectacles, text. 1.0
No full textThe dataset contains the primary sources at the core of the Horizon 2020 ANIMATE project, namely the 12 albums created by Giovanni Tommaso Borgonio for spectacles performed at the Savoy court of Christine of France between 1640 and 1660. The 8 albums held at the Biblioteca Nazionale Universitaria di Torino are included in the dataset as .pdf format files (these albums have been digitised and are available free of charge as .pdf files in low resolution on the "Internet Culturale" portal, https://www.internetculturale.it/). Due to copyright restrictions, the 3 albums held at the Biblioteca Reale di Torino and the 1 held in a private collection are not preserved and made available in the dataset as .pdf. As an alternative, a bibliography of these four albums is provided with detailed information in tabular form (i.e. typology of the performance, author, title, place, date, author of the album, location of the album, call number, number of sheets, language, date of the album, extant printed report, extant sheet music). This .cvs file also contains the bibliographic details of the 8 albums held in at the Biblioteca Nazionale Universitaria di Torino. The dataset contains also a .cvs with the information regarding the sources (i.e. 17th-century account books) examined at the Archivio di Stato di Torino (location, section, archival fonds, header, volumes, dates, subject of the information) and a .cvs with the information regarding the sources (i.e. 17th-century printed reports for festivals staged across the Duchy of Savoy) examined at the Archivio Storico di Stato della Città di Torino (location. collection, box, date and place, title)
Fatherland, counterrevolution, romanticism. Xavier de Maistre and his lost Savoy
No full textThe literary works of Xavier de Maistre, savoyard writer and painter of the turn of 18th and 19th centuries, are strictly connected to the era in which the author has lived, particularly in the way of considering concepts of fatherland and nature, landscape. A trauma, related to the invasion of French revolutionary forces in Savoy pushed de Maistre into the counter‐ revolutionary camp and stimulated narrative reflections upon the themes of homeland, nature, and political change. His works, such as Voyage autour de ma chambre and Expédition nocturne autour de ma chambre, written between 1794 and 1825, show a particular evolution of the author’s point of view, starting from a classical, objectifying approach, inherited form the Enlightenment, to transform finally towards a subjective, individual scope, specific to the romanticism. To discover the motors and instruments of this evolution of the author’s mind is the object of this paper
Weight stigmatization experiences and psychological adjustment: the role of dysfunctional cognitions
No full textWeight-based stigmatization experiences include negative social consequences that overweight and obese individuals endure. Although there is evidence that weight-based stigmatization places one at risk for developing internalizing symptoms and disordered eating symptoms, further research is needed to establish causality and determine what underlying processes are involved in these pathways. This dissertation research investigates the hypothesis that dysfunctional cognitions mediate the relation between weight-based stigmatization experiences and undergraduates’ psychopathology and weight-biased behavior. Results indicate that weight-based stigmatization experienced through exposure to weight-biased media is associated with dysfunctional eating-related cognitions and that exposure to weight-biased media is associated with weight-biased behavior. Evidence did not support the hypothesis that dysfunctional eating-related cognitions would account for the relation between weight stigmatization experiences and disordered eating symptoms. Instead, weight-related stereotypes emerged as a moderator of this relation. This research has implications for prevention of disordered eating and weight-biased behavior and interventions with individuals who suffer weight-based stigmatization.Ph.D.Includes bibliographical referencesIncludes vitaSarah Savo
FCNC in SUSY theories
No full textURL: http://www-spht.cea.fr/articles/T96/003 (sur invitation). FCNC dans les théories SUSY http://fr.arxiv.org/abs/hep-ph/9601225International audienceRecent work on flavour changing neutral current effects in supersymmetric models is reviewed. The emphasis is put on new issues related to solutions to the flavour problem through new symmetries: GUTs, horizontal symmetries, modular invariances
Some Supersymmetric Flavour Problems
No full textURL: http://www-spht.cea.fr/articles/t99/155/ Quelques problèmes supersymétriques reliés aux saveursCours, Seminaire, Ecole, School NATO advances Cargèse Summer School on Hierarchies of Fermion Masses and Gauge InteractionsSome Supersymmetric Flavour Problems | Quelques problèmes supersymétriques reliés aux saveur
- …
