A well-known fact about positional election procedures is that its ranking of m alternatives can change when some of the alternatives are removed from consideration—given a positional procedure on each of 2, 3, …, m alternatives and a collective preference order for each distinct subset of the m alternatives. Saari has established that with few exceptions, we can find a voter profile for which the collective preference order for each subset under the according positional procedure is the one given. However, Saari's results do not quantify the likelihood of finding such voter profiles. For small numbers of alternatives, William Gehrlein developed a statistical model to explore the probabilities that particular collective preference orders on subsets of alternatives can occur for large electorates. One goal of this research is to determine whether changes in the collective preference order as alternatives are removed can be considered to be the norm or an outlier for positional procedures. This dissertation extends the research headed by Gehrlein in two directions. One, I generalize his statistical model to explore probabilities for iterated election procedures. Gehrlein's model previously produced results only for three alternatives and in limited cases for four alternatives. I have extended this model to produce results for up to five alternatives, including analysis of instant-runoff voting and runoff elections. Two, Gehrlein's model required specific conditions on the probability distribution of individual voter preferences across the population. I relax this assumption so that for any probability distribution of individual voter preferences across the population, I can explore the probability that a collective preference order is inconsistent with the outcomes when alternatives are removed. These results provide a foundation for discussing the impact of removing alternatives on elections across all large electorates. I also apply these results to two recent United States elections wherein a third-party candidate received a significant share of the votes: the 1992 U.S. Presidential election and the 1998 Minnesota Gubernatorial election. Overall, my research will suggest that as the number of alternatives increases, the likelihood of finding changes in the collective preference order as alternatives are removed will approach one.

Les interactions plante-virus diffèrent des autres interactions plante-pathogènes du fait de la nature des virus qui sont des parasites intracellulaires obligatoires. Plus spécifiquement, l’interaction haricot commun (Phaseolus vulgaris L.)-Bean pod mottle virus (BPMV) a été étudiée en mettant l’accent à la fois sur la résistance de la plante mais aussi sur la virulence du virus dans l’objectif de mieux comprendre et d’identifier les facteurs intervenant dans le dialogue moléculaire entre plante et virus. Ces deux partenaires interagissent selon le modèle «gène-à-gène» de Flor. 1) Côté plante, nous avons identifié un gène de résistance dominant vis-à-vis du BPMV chez BAT93, le gène R-BPMV. Ce gène est localisé à l’extrémité du chromosome Pv02, dans la région du locus I, un locus de résistance multi-parasitaire vis-à-vis de différents virus, bactérie et champignon. La cartographie fine du gène R-BPMV suivie du séquençage de la région à partir d’un contig de clones BACs chez BAT93 a permis d’identifier des séquences codant pour des protéines NB-LRR qui pourraient correspondre au gène R-BPMV. Des études de microsynténie et de phylogénie ont été réalisées afin de mieux comprendre l’évolution des gènes présents dans cette région. L’étude au niveau cellulaire du phénotype associé à la résistance a permis de montrer que le gène R-BPMV bloque le mouvement de cellule à cellule du virus et que le phénotype associé est température-dépendant. 2) Côté virus, le clonage de toutes les ORFs du BPMV associé à des expériences d’agroinfiltration sur P. vulgaris et Nicotiana benthamiana ont permis d’identifier deux facteurs viraux importants dans le dialogue moléculaire plante-virus : la protéine VPg du BPMV correspond à la protéine d’avirulence agissant en interaction avec le produit du gène R-BPMV dans le cadre du modèle «gène-à-gène», et l’ARN polymérase ARN-dépendante virale correspond à un suppresseur de silencing à effet faible. 3) A ce jour, la transformation génétique stable n’est pas applicable en routine chez les légumineuses. Un objectif de la thèse est de développer des outils de validation fonctionnelle pouvant s’appliquer à des gènes d’intérêt agronomique, dont des gènes de résistance aux maladies. L’approche VIGS basée sur un vecteur viral dérivé du BPMV, déjà utilisée chez le soja, a ainsi été adaptée sur haricot et pois (Pisum sativum), une légumineuse économiquement importante en Europe. / Plant-virus interactions differ from other plant-pathogen interactions because viruses are obligate intracellular parasites. More specifically, common bean (Phaseolus vulgaris L.)-Bean pod mottle virus (BPMV) interaction was studied by focusing both on the plant resistance and on the virus virulence in order to highlight and identify factors involved in the molecular dialog between plant and virus. These two partners interact according to the “gene-for-gene” model described by Flor. 1) On the plant side, we identified a dominant resistance gene against BPMV in cv. BAT93, the R-BPMV gene. This gene is located at one end of chromosome Pv02 in the I locus region, a multi-parasitic resistance locus involved in resistance to different viruses, bacteria and fungi. Fine mapping of R-BPMV followed by sequencing of the region from a BACs contig in BAT93 allowed us to identify sequences encoding NB-LRR proteins that could correspond to R-BPMV. Microsynteny and phylogeny studies were performed to understand the evolution of genes present in this region. When resistance phenotype was studied at the cellular level, we found that R-BPMV blocks BPMV cell-to-cell movement and that resistance phenotype is temperature-dependent. 2) On the virus side, cloning of all BPMV ORFs in association with agroinfiltration assays in P. vulgaris and Nicotiana benthamiana allowed us to identify two important factors involved in plant-virus molecular dialog: the BPMV VPg acting as an avirulence factor in interaction with the product of R-BPMV in the “gene-for-gene” model, and the viral RNA-dependent RNA polymerase that corresponds to a weak RNA silencing suppressor. 3) To date, stable genetic transformation is not routinely feasible in legumes. One objective of this thesis was to develop news tools for functional validation studies for genes of agronomic interest, including disease resistance genes. The VIGS approach based on the viral BPMV vector, first used in soybean, was adapted to common bean and pea (Pisum sativum), a legume species of high economic importance in Europe.

La molécule de l’éthylène (C2H4) est le régulateur principal du sexe chez les cucurbites. Essentiellement, l’éthylène est connu pour son rôle promoteur dans le développement des carpelles et un rôle inhibiteur des étamines dans les fleurs du melon. L’interaction entre les biosynthétique gènes de l’éthylène (CmACS7, CmACS11, et CmACO3) et le facteur de transcription CmWIP1 détermine différentes formes du sexe chez le melon. Le rôle de ces gènes est bien étudié chez le melon. Cependant, le mécanisme qui contrôle l’initiation et la coordination de formation des étamines et des carpelles dans la fleur reste ambigu. En reposant sur l’importance de l’éthylène dans l’expression du sexe chez le melon, j’ai focalisé sur l’identification des gènes impliqués dans la voie signalisation éthylène-sexe. Au cours de la thèse, le criblage des mutants altérés dans la réponse à l’exogène éthylène nous facilitait d’identifier des nouveaux gènes impliqués dans la détermination du sexe chez la famille de Cucurbitacée. Pendant ma thèse j’ai isolé plus de 10 mutants insensibles à l’éthylène de différentes populations du melon. Deux mutants ont été isolés de deux populations monoïques indépendantes. Ces deux mutants provoquent une transition partielle et complète au melon andromonoïque dans la génération M2, respectivement. Un de ces deux mutants a été identifié et caractérisé. Deux autres mutants gsn106 et vat233 ont été criblés de deux populations andromonoïque, provoquent une transition complète et partielle à androïque melon, respectivement. En utilisant le séquençage à haute débit et les analyses génétiques j’ai essayé de cloner et caractériser ces gènes mutants. Par ailleurs, des autres mutants insensibles à l’éthylène sont en cours d’être phénotypes pour le phénotype du sexe. L’isolation et caractérisation des nouveaux gènes impliqués dans le déterminisme du sexe nous aidera pour mettre en place un model clair explant comment le sexe est contrôlé chez les plantes. / Ethylene (C2H4) is an important phytohormone in plants and the main sex regulator in the family Cucurbitaceae. As known, the ethylene promotes the carpel development and inhibits the stamens in the melon flower (Cucumis melo L.). The interplay of the biosynthesis genes (CmACS7, CmACS11, et CmACO3) and the transcription factor CmWIP1 generates different sexual forms in melon. The role of these genes in the sex expression is well studied. However, the mechanism that controls the initiation and coordination of stamen and carpel development in the flower remains ambiguous. Based on the importance of the ethylene in the sex determination, I aimed to isolate novel genes involved in the pathway ethylene-sex in the melon (Cucumis melo L.). For this purpose, I used the response to exogenous ethylene in the etiolated seedlings (known as the triple response phenotype) to isolate ethylene-insensitive mutants. During my thesis I isolated more than 10 ethylene-insensitive mutants from six EMS-mutagenised melon populations. Some of these mutants induced changes in the sex expression of the melon. . Two mutants were isolated from two independent monoecious populations (female and male flowers on the same plant) and induced a partial and a complete sexual transition to the andromonoecious melon in the second generation M2, respectively. One of them was cloned and characterized using Omics tools. Two other mutants (gsn106) and (vat233) screened from two independent andromonoecious melon (bisexual and male flowers on the same plant) populations, induced complete and partial sexual transitions into androecy (only male flowers), respectively. Using Next-Generation Sequencing (NGS) and the genetic analysis, we are trying to clone and characterise these mutants (gsn106) and (vat233). In the same way, we continue to observe others promising ethylene-insensitive mutants (vat306, vat175, and vat230) for the sex phenotype. The isolation and characterisation of novel genes involved in the sex determination will permit to provide a new and clear model explains of the sex determination mechanism in plants.

Author: Hirofumi Sageshima Title: Electromyographic analysis of the cervical flexion during different body positions Aim and Purpose: The aim of this study is to find out the muscle activity and the muscle coordination during cervical flexion in 3 different positions - standing, sitting and supine - with electromyographic analysis in young healthy adults. Methods: This study compared superficial cervical flexor muscles activity during conventional cervical flexion to 20 volunteered healthy subjects with surface electromyography (EMG). The activity of 5 paired muscles - sternocleidomastoid, scalenus, suprahyoid, infrahyoid and superior part of trapezius - were measured when they performed 15-time cyclic cervical flexion-extension from neutral to maximum cervical flexion on 3 different positions - sitting, standing and supine - in random order. Results: Significantly higher EMG amplitude was detected from all measured muscles on supine position than sitting and standing (p < 0.05). It was also confirmed that muscle activation pattern was different according to positions; its amplitude on supine reach the peak in the earlier phase of movement, while it was delayed on 2 other positions (p < 0.05). In terms of onset, all cervical flexors activated together in the very beginning. However, on sitting and...

Geospatialt data kan samlas in på en mängd olika sätt, där laserskanning och flygfotografering är några av dem. Genom de antal tekniker som möjliggör insamlingen av geospatialt data existerar lösningar såsom användning av Global Navigation Satellite System, Light Detection and Rangning, Structure-from-Motion m.fl. I och med dessa teknologier kan datamängder som höjdmodeller, punktmoln och ortofoton produceras. Inom Sverige finns det en mängd datakällor som kommuner kan använda sig av, där bl.a. Lantmäteriet är en av dem. Då det finns en risk av en fördröjning i uppdatering av data går kommuner runt om i landet potentiellt miste om uppdaterat data. En av de möjliga lösningarna för att få tillgång till uppdaterat data är att kommuner själva samlar in data. I och med denna studie undersöks om drönare är ett lämpligt alternativ till andra datakällor. Vid undersökningen utvärderades skillnader mellan datamängder som var georefererat med markkontrollpunkter och utan i relation till Lantmäteriets data. De datamängder som skapades baserat på drönarens flygbilder varierade i kvalité i kontrast till de inställningar som användes i Agisoft. Genom beräkning av kvadratisk medelavvikelse, differens, punktavstånd och punktäthet identifierades skillnader i kvalité i de skapade datamängder: höjdmodeller, punktmoln och ortofoton. Resultaten av detta visade att data som var georeferat med markkontrollpunkter hade högst lägesnoggrannhet och högre än Lantmäteriets data. Ytterligare hade de skapade datamängder i denna studie högre upplösning (ortofoton &amp; höjdmodeller) samt punktäthet och punktantal (punktmoln). Genom resultaten visar en av slutsatserna i denna studie att lämpligheten för att samla in data på eget bevåg av tillräcklig god kvalité är möjlig och att detaljgraden i relation till Lantmäteriets data är högre / Spatial data can be collected in a variety of ways, where laser scanning and aerial photography are some of them. With the use different technologies such as Global Navigation Satellite System, Light Detection and Ranging, Structure-from-Motion and more, enables the collection of spatial data. With these technologies, datasets such as elevation models, point clouds and orthophotos can be produced. Within Sweden, there are several data sources that municipalities can use, where Lantmäteriet is one of them. Since there is a risk of a delay in updating data, municipalities around the country are potentially losing out on updated data. The possible solutions for accessing updated data is that municipalities themselves collect data. This study will investigate whether drones are a suitable alternative to other data sources. In the survey, differences between data sets that were georeferenced with ground control points and without in relation to data from Lantmäteriet were evaluated. The amounts of data that were created based on the drone's aerial images varied in quality in contrast to the settings used in Agisoft. By calculating root mean square error, difference, point distance and point density, differences in quality were identified in the datasets created: elevation models, point clouds and orthophotos. The results showed that data georeferenced with ground control points had the highest positional accuracy and higher than the data from Lantmäteriet. Further, the datasets created in this study had higher resolution (orthophotos &amp; elevation models) as well as point density and number of points (point clouds). Through the results, one of the conclusions of this study shows that the suitability for collecting data on municipalities own accord of sufficiently good quality is possible and that the degree of detail in relation to data from Lantmäteriet is higher.

Hybrid sterility is one of the reproductive isolation mechanisms restricting gene flow between the related species and leading to speciation. PR domain containing 9 (Prdm9), the only known vertebrate hybrid sterility gene, determines the sites of programmed DNA double-strand breaks (DSBs) and thus specifies hotspots of meiotic recombination but in hybrids between two mouse subspecies causes failure of meiotic chromosome synapsis and hybrid male sterility. In the present study on sterile hybrids, the five smallest autosomes were more prone to asynapsis. To manipulate with the synapsis rate, random stretches of consubspecific homology were inserted into several autosomal pairs. Twenty seven or more megabases of consubspecific sequence fully restore synapsis in a given autosome. Further, at least two symetric DN double-strand breaks per chromosome were necessary for successful synapsis. Moreover, F1 hybrids had sperm when synapsis was rescued in at least three of four segregating chromosomes. To verify the assumption of a lack of symmetric DSBs in meiotic chromosomes of sterile males the chemotherapeutic drug cisplatin was used to induce exogenous DNA DSBs. Cells treated with 5 mg/kg and 10 mg/kg of cisplatin showed increased number of DSBs monitored by immunostaining of RPA and DMC1 sites and...

[pt] Vídeos se tornaram a principal mídia para a comunicação, com um volume massivo de dados criado a cada segundo. Conseguir entender essa quantidade de dados de forma automática se tornou importante e, por conseguinte, métodos de video understanding são cada vez mais necessários. Uma tarefa crucial para o entendimento de vídeos é a classificação e localização no tempo de diferentes ações. Para isso, a segmentação de ações precisa ser realizada. Segmentação de ações é a tarefa que consiste em segmentar temporalmente um vídeo, classificando cada quadro com alguma ação. Neste trabalho, é proposto um método de segmentação de ações que não requer análise prévia do vídeo e nenhum dado anotado. O método envolve a extração de embeddings espaço-temporais dos vídeos com redes de aprendizado profundo pré-treinadas, seguida por uma transformação realizada por um codificador posicional e pela aplicação de um algoritmo de grupamento em que cada cluster gerado corresponde a uma ação diferente. Os experimentos realizados demonstram que o método produz resultados competitivos nos conjuntos de dados Breakfast e Inria Instructional Videos. / [en] The rise of video content as the main media for communication has been creating massive volumes of video data every second. The ability of understanding this huge quantities of data automatically has become increasingly important, therefore better video understanding methods are needed. A crucial task to overall video understanding is the recognition and localisation in time of dierent actions. To address this problem, action segmentation must be achieved. Action segmentation consists of temporally segmenting a video by labeling each frame with a specific action. In this work, we propose a novel action segmentation method that requires no prior video analysis and no annotated data. Our method involves extracting spatio-temporal features from videos using a pre-trained deep network. Data is then transformed using a positional encoder, and finally a clustering algorithm is applied where each cluster presumably corresponds to a dierent single and distinguishable action. In experiments, we show that our method produces competitive results on the Breakfast and Inria Instructional Videos dataset benchmarks.

Starting from the experience of watching the role of the teacher turn into a bureaucrat, this work examines the factors dominating current education systems. It identies Human Capital Theory as a main inuence on policy and prioritization decisions in education. A central tension that results from this inuence is the fact that economic theory assumes consumer sovereignty, which dismisses many ideological and political aspects of education as paternalism. This inherent conict between economic theory and educational theory leads to an overemphasis on positional goods, goods that are limited by scarcity, over nonpositional goods, such as knowledge. Educating for positional goods is contrasted to the Spinozistic ideal of education for freedom and for increased conatus. It is then discussed how current assessment practices play a central role in cementing the current hegemony in education as governed by economic concerns. Assessment practices are found to both contribute to current trends in education as well as stabilizing the system overall making transformative change less likely. This is more specically illustrated by two particular assessment practices, namely standardized testing and assessment for learning. It concludes with a search for possible paths to meaningful change in education.

No description available.

With the continued proliferation of access and the usage of the internet, the field of web learning is continuously growing in order to automate and improve parts of our experience on the web. Research in web learning has often lagged behind its counterparts in Natural Language Processing (NLP), novel methods often reach adoption in web learning research with a delay. Web pages are more complex in both content and structure, as they are semi- structured documents divided into sections, often containing a combination of images, text, and markup. For humans, this is not difficult to understand, as we are familiar with the structure of web pages and in fact are often aided by the styling and markup of the pages. However for machine learning algorithms, this structure and mixture of content poses several challenges which are not similar in nature to comparable documents in NLP problems. Transformer models have shown significant performance gains on a multitude of tasks ranging from NLP to image processing. This thesis studies the usage of alternative and novel approaches to encoding positional information of nodes in a HyperText Markup Language (HTML) Document Object Model (DOM) tree in order to enable effective use of transformer models on web page data. The problem studied was a HTML element classification problem, specifically the task of extracting product data from a product web page. Three positional encodings for tree structured data were studied: Breadth First Search (BFS), Depth First Search (DFS), and novel tree positional encodings. These encodings resulted in 3 trained transformer models which were compared to a baseline transformer model trained with no positional encoding in order to measure the change in performance that the encodings produced. The analysis of the results show that the BFS and DFS encodings increased model performance across all measured metrics (precision, recall, f1-score, accuracy) by up to 1% in absolute performance. The novel tree positional encodings resulted in worse model performance across all metrics measured. The results show that transformers benefit from certain tree positional encodings of the HTML elements, and further research should be done to see how these positions can be effectively encoded for transformer models to process web pages. / Med den fortsatta spridningen av åtkomst och användningen av internet växer området för webbinlärning kontinuerligt för att automatisera och förbättra vår erfarenhet på nätet. Forskning inom webbinlärning har ofta släpat efter dess motsvarigheter inom NLP, nya metoder når oftast webbinlärningsforskning med försening. Webbsidor är mer komplexa i både innehåll och struktur än text dokument, eftersom de är halvstrukturerade dokument indelade i sektioner, som ofta innehåller en kombination av bilder, text och stil. För människor är detta inte svårt att förstå, eftersom vi är bekanta med strukturen på webbsidor och faktiskt ofta får hjälp av utformningen och uppdelningen av sidorna. Men för maskininlärnings algoritmer är strukturen och blandningen av innehåll en utmaning som inte liknar jämförbara dokument i NLP. Transformer modeller har visat flera prestandaförbättringar på en mängd uppgifter som sträcker sig från NLP till bildbehandling. Denna uppsats studerar användningen av alternativa och nya metoder för kodning av positionsinformation för noder i ett HTML DOM träd för att möjliggöra effektiva användningen av transformer modeller på webbsidadata. Problemet som studerades var ett HTML elementklassificeringsproblem, specifikt uppgiften att extrahera produktdata från en produktsida. Tre positionskodningar för trädstrukturerade data har studerats: BFS, DFS och träd positionskodningar. Dessa kodningar resulterade i tre tränade transformer modeller som jämfördes med en modell tränad utan någon positionskodning för att mäta förändringen i prestanda som kodningarna producerade. Analysen av resultaten visar att BFS och DFS kodningarna ökade modellprestanda över alla uppmätta mått (precision, accuracy, f1-score) med upp till 1% i absolut prestanda. De nya trädpositionskodningarna resulterade i sämre modellprestanda över alla mått mätt. Resultaten visar att transformer modellererna drar nytta av vissa trädkodningar för HTML-elementen, och ytterligare undersökningar bör göras för att se hur dessa positioner av datan effektivt kan kodas för transformer modeller för att bearbeta webbsidor.