Context: Determining attribute correspondence is the most important, time consuming and knowledge intensive part during databases integration. It is also used in other data manipulation applications such as data warehousing, data design, semantic web and e-commerce. Objectives: In this thesis the aim is to investigate how to find the types of correspondence between the attributes of heterogeneous datasets when schema design information of the data sets is unknown. Methods: A literature review was conducted to extract the knowledge related to the approaches that are used to find the correspondence between the attributes of heterogeneous datasets. Extracted knowledge from the literature review is used in developing an instance based approach for finding types of correspondence between the attributes of heterogeneous datasets when schema design information is unknown. To validate the proposed approach an experiment was conducted in the real environment using the data provided by the Telecom Industry (Ericsson) Karlskrona. Evaluation of the results was carried using the well known and mostly used measures from information retrieval field precision, recall and F-measure. Results: To find the types of correspondence between the attributes of heterogeneous datasets, good results depend on the ability of the algorithm to avoid the unmatched pairs of rows during the Row Similarity Phase. An evaluation of proposed approach is performed via experiments. We found 96.7% (average of three experiments) F-measure. Conclusions: The analysis showed that the proposed approach was feasible to be used and it provided users a mean to find the corresponding attributes and the types of correspondence between corresponding attributes, based on the information extracted from the similar pairs of rows from the heterogeneous data sets where their similarity based on the same common primary keys values.

Cette thèse est organisée en deux parties. La première partie, qui porte sur la morphosémantique de l’adjectif, est divisée en trois chapitres. Le premier est consacré à l’étude de l’expression de la qualité en berbère. Nous avons commencé, d’abord, ce premier chapitre par un bilan des travaux effectués sur l’adjectif en berbère. Puis, nous avons présenté les différentes procédures de l’expression de la qualité, à part l’adjectif, à savoir les complexes qualifiants, les verbes de qualité et les formes participiales. Dans le deuxième chapitre, nous avons étudié certains aspects morphologiques de l’adjectif, notamment ses procédures de formation, ses schèmes et ses formes. Nous avons effectué également, au sein de ce deuxième chapitre, une comparaison, d’une part, entre l’adjectif kabyle et l’adjectif dans d’autres parlers berbères et d’autre part, entre l’adjectif en berbère et l’adjectif dans certaines langues chamito-sémitiques. Nous nous sommes efforcés de définir, dans le troisième chapitre, les limites de la classe de l’adjectif. Nous y avons abordé les questions de la proximité des substantifs avec les adjectifs et le parallélisme de ces derniers avec les formes verbales. La deuxième partie de ce travail, quant à elle, touche à la syntaxe de l’adjectif. Nous y avons étudié comment l’adjectif s’insère dans les propositions, avec ses fonctions principales, celle d’épithète et celle de prédicat. Ce dernier chapitre porte également sur les différents modifieurs que les adjectifs admettent, les éventuels compléments dont ils peuvent être pourvus et les coordinations qui leur sont permises soit entre eux, soit avec les éléments qui les accompagnent au niveau syntagmatique. / This thesis is divided into two parts. The first part, which deals with the morpho-semantics of the adjective, is divided into three chapters. The first one studies the expression of quality in Berber. We start by making the inventory of the studies devoted to the adjective in Berber. Then, we introduce the different ways of expressing quality beside adjectives: complex qualifiers, verbs of quality and participial forms. In the second chapter, we study some morphological aspects of the adjective, particularly its formation process, its schemas and forms. Moreover, we draw a comparison between the adjective in Kabyle and in other Berber languages on the one side, and between the adjective in Berber and in some Hamito-Semitic (Afroasiatic) languages on the other. In the third chapter, we try to define the limits of the adjectival class. We tackle the question of the proximity of substantives with adjectives and the parallelism of the latter with verbal forms. The second part of this study deals with the syntax of the adjective. We study the way the adjective fits into propositions, in attributive and predicative uses. This last chapter also studies the different modifiers which adjectives allow the possible complements they can accept, and the coordinations they are likely to have among themselves or with the elements which are contiguous to them at the syntagmatic level. / Amahil-agi, yebḍa ɣef sin yeḥricen. Aḥric amezwaru, yewwi-d ɣef telɣasnamekt. Deg-s kraḍ n waxfiwen. Deg yixef amezwaru, yella-d wawal ɣef uwenni n tɣara di tmaziɣt. Di tazwara, newwi-d ameslay ɣef wayen yettwarun ɣef wefris di tmaziɣt. Syin ɣer-s, nesken-d amek i d-yettili uwenni n tɣara di teqbaylit, ama s wuddisen n tɣara, ama s wemyag n tɣara, ama s umaɣun. Deg yixef wis-sin, nesleḍ kra n temsal yerzan talɣa n wefris di teqbaylit, am iberdan swayes yettwalɣu, am tseqqal-is, am talɣiwin-is. Nerna nemzazal, di tama ger wefris di teqbaylit d wefris di tentaliyin nniḍen n tmaziɣt, di tama nniḍen ger wefris di tmaziɣt d wefris di tutlayin tiḥami-samiyin nniḍen. Ma deg yixef wis-kraḍ, nemmeslay-d ɣef tilas n wesmil ufris. Nefka-d ayen i t-izdin d wayen i t-isemgarden d weɣrik. Nerna-d diɣ ayen i t-izdin d wayen i t-isemgarden d telɣiwin timyaganin. Deg weḥric wis-sin, yella-d wawal ɣef tseddast n wefris di teqbaylit. Nutlay-d ɣef wemḍiq n wefris di tsumert, ama ticki yella d anemlaway, ama ticki yella d inemmi. Deg yixef-agi aneggaru, newwi-d ameslay ɣef imerna d yesmmuden yettawi wefris akked usmmezdi i d-yettilin ger yefrisen neɣ ger wefris d iferdas ideg yeddukkul deg wusniy.

Current marketing discourse seems to be less and less about communicating to customers, and more and more about communicating with customers. Concurrently, traditional news companies are in a bit of a crisis where their business models are failing to compete with the growing colossus of “free” news that circulates the web. In the aftermath of all this, native advertising – an umbrella term for advertisements that look like editorial content – has emerged like a panacea. But while native advertisements might be effective where traditional advertisements are not, they also create a problematic ambiguity for the consumer – what is marketing and what is news? This thesis set off to investigate whether consumers perceive native advertisements as advertisements and how that might affect their trust towards the advertiser (company) and platform (news site). A questionnaire with a slightly experimental touch was constructed from various theoretical focal points; 207 respondents were exposed to three different types of native advertisements and then answered statements about how they perceived them on a 7-point Likert scale. The majority of the respondents considered all three of the native advertisements as sponsored content, with regard to some notable disparities between the examples. However, not all of them did; in one case, as much as 27 % stated they did not perceive the advertisement as sponsored content at all. There was also a unanimous preference towards traditional advertisements, particularly among those who felt they had a hard time distinguishing native advertisements from editorial content, and frequent occurrence of sponsored content on a news site weakened its general credibility among the respondents, as well as their trust in both advertiser and platform. This study is more exploratory than conclusive and should be interpreted as such, but from these results it seems that stakeholders should tread a lot more carefully when using native advertisements compared with traditional advertisements – particularly considering how they have become an industry norm in an almost dogmatic fashion. The consequences may not be as direct as a wound up CTR-ratio or increased advertising revenue, but trust is simply a too valuable currency to gamble with in an era filled with ‘post-truth’ and ‘fake news’. / Aktuell marknadsföringsdiskurs verkar handla mindre och mindre om att kommunicera till konsumenter, och mer och mer om att kommunicera med konsumenter. Parallellt med det har de mer traditionella nyhetsbolagen hamnat i något av en kris efter att deras affärsmodeller misslyckats med att konkurrera med den växande kolossen av ”gratis” nyheter som cirkulerar på webben. I efterdyningarna av allt detta har native advertising – ett paraplybegrepp för annonser som liknar redaktionellt innehåll – uppstått som en slags universallösning. Men medan native advertisements framstår effektiva där traditionella annonser inte gör det så uppstår även en problematisk ambiguitet för konsumenten – vad är marknadsföring och vad är nyheter? Denna uppsats tog avstamp i att undersöka om konsumenter uppfattar native advertisements som annonser och hur det påverkar deras förtroende gentemot annonsör (företag) och plattform (nyhetssida). En enkät med en något experimentell touch konstruerades utifrån olika teoretiska infallsvinklar; 207 respondenter exponerades för tre olika varianter av native advertisements, varefter de fick besvara påståenden om hur de uppfattade dem på en sjugradig Likert-skala. Majoriteten av respondenterna ansåg alla tre native advertisements vara sponsrade, med hänsyn till viss diskrepans mellan exemplen. Emellertid var det långtifrån alla som gjorde det; i ett fall var det så många som 27 % som menade att de inte uppfattade annonsen som sponsrat innehåll. Det var också en övervägande preferens gentemot traditionella annonser, särskilt bland de som tyckte det var svårare att urskilja native advertisements från redaktionellt innehåll, och en hög förekomst av sponsrade artiklar på en nyhetssida försämrade inte bara dess generella trovärdighet bland respondenterna, utan också deras förtroende för såväl annonsör som plattform. Studien är mer explorativ än fastställande och bör tolkas därefter, men utifrån dessa resultat framstår det som att alla intressenter behöver träda mycket mer varsamt när de använder sig av native advertisements istället för traditionella annonser – i synnerhet med tanke på hur de blivit en branschnorm på ett nästan dogmatiskt vis. Konsekvenserna kanske inte är lika direkta som en uppskruvad CTR-ratio eller ökade annonsintäkter, men förtroende är helt enkelt en alldeles för värdefull valuta för att gambla med i en era fylld av ’post-sanning’ och ’alternativa nyheter’.

Doctor of Philosophy / Department of Psychology / Laura A. Brannon / Two studies address the serious problem of college student binge drinking. Both studies identify factors that improve the effectiveness of public service announcements (PSAs) encouraging responsible drinking presented through a website simulation. Study 1 tested four levels of Message Personalization (i.e., extent to which the PSA targets important aspects of the individual's personality) by comparing the effectiveness of messages matched to the person's Big Five personality traits, their actual self-schema, their ideal self-schema, or a non-personalized control message. Matching to actual self-schema has been found to be effective in past research. However, it was expected that the more thoroughly personalized the message, the more effective it would be. Results revealed that in no instance was the most thoroughly personalized condition (Big Five matched) or the alternate way of matching to schema (ideal self-schema) more effective than the actual self-schema matching. When designing PSAs, there appears to be a threshold of personalization. Research related to testing PSAs discouraging binge drinking should continue to pursue self-schema matching rather than the more complicated Big Five matching. Study 2 tested Person Matching (i.e., whether the PSA matches the person's self-schema type or not) and two types of Context Matching (i.e., whether the PSA matches the Topic or Values of the message context) to determine their relative influence on the effectiveness of the PSA. It was expected that PSAs matched to any of these factors would be more effective than messages not matched, and that Person Matching would be more influential on the PSAs effectiveness than the two types of context matching. Person Matching reduced intentions to drink while staying in/home, but Topic Matching reduced intentions to drink when going out, suggesting that different factors are important for PSAs targeting drinking behavior in different locations. The interaction of Topic Matching and Values Matching indicate that the PSA should not match the message context too closely. Again, there appears to be a matching threshold; increasing the number of factors the message matches does not increase message effectiveness, possibly because it makes the message too redundant with the webpage content.

Cette thèse interroge les images dans le parcours de soin de femmes touchées par un cancer du sein. Nous questionnons de façon spécifique les images ayant trait à la chirurgie curative du cancer du sein, images représentant la transformation du corps de la femme. Notre objectif est de comprendre les usages des images dans la consultation chirurgicale et de saisir leur interaction avec l’expérience de la maladie vécue par les patientes. Nous nous appuyons sur l’approche des représentations sociales pour appréhender la notion d’image selon deux acceptions. D’une part, l’image renvoie à l’iconographie ; ce qui nous intéresse alors est l’objectivation graphique, forme d’objectivation qui ne passe pas par les discours mais par la transformation des idées en entités concrètes dans des objets graphiques réels, comme l’iconographie circulant dans différents contextes. D’autre part, la notion d’image nous sert à désigner la dimension figurative des représentations sociales. Sous cette signification, l’image renvoie à l’aspect concret des représentations sociales, issu du processus d’objectivation. Ces deux acceptions de la notion d’image sont mises en lien avec la dimension expérientielle et les processus de genèse et de transformation des représentations sociales. Les méthodes d’enquête que nous avons mises en place sont qualitatives et s’inscrivent dans des lieux de soin. Nous nous appuyons sur un plan méthodologique reposant sur le principe de la triangulation des méthodes, qui permet d’atteindre différentes facettes de l’objet, dans différents contextes. Dans notre étude, la méthode d’observation de consultations chirurgicales est éclairée par des entretiens et focus groups avec des patientes.Les observations de consultations permettent de relever les types d’images utilisés dans ce contexte et les usages qui en sont faits : photographies et schémas sont principalement employés pour représenter la chirurgie du cancer du sein, chaque type d’image mettant en scène le corps selon des caractéristiques singulières. Si les patientes commentent peu ces images au cours des consultations, entretiens et focus groups nous donnent accès à leur perspective : photographies et schémas n’entrent pas en interaction de la même manière avec l’expérience des patientes et ne renvoient pas aux mêmes univers représentationnels. / This thesis questions the images in the treatment of breast cancer. More particularly, we question the images related to breast cancer curative surgery representing the transformation of the woman body. Our objective is to understand the use of the images in surgical consultation and to seize their interaction with the experience of the disease by the patients. We rely on the approach of social representations to comprehend the concept of image according to two different acceptances. On the one hand, the image refers to iconography; therefore, what interests us is the graphical objectification, which does not imply speeches but the transformation of ideas into concrete entities in real graphical objects such as the iconography available in different contexts. On the other hand, the concept of image is used to refer to the figurative aspect of social representations. Under this meaning, the image refers to the concrete aspect of social representations emanating from the objectification process. These two acceptances of the concept of image are linked to the experience dimension and to the genesis and transformation processes of social representations. We used qualitative investigating methods and worked in care centers. Our methodology is based on triangulation, which enabled us to assess different aspects of the object in different contexts. The method used to observe surgical consultations in our study was enlightened by interviews and focus groups held with patients.The consultation observations enabled us to highlight the types of images used in this context and how they are used: photographs and schemas are mainly used to represent breast cancer surgery as each type of image shows the body according to specific features. While the patients do not comment much these images during the consultations, they make more comments during interviews and focus groups, showing us their perspectives: photographs and schemas do not interact the same way with the experience of the patients, and do not refer to the same representational spheres.

This thesis introduces an orienting mechanism and control system for the purpose of eye tonometry. The design is based on a 3RRR spherical parallel manipulator architecture. The end-effector is mounted with a triad of force sensing elements. Presented in this paper is a unique approach to force control based on an artificial neural network. The mechanism generates movements to collect data about its tactile environment ultimately generating a path to the force sensors' equilibrium point.

The aim of this thesis is to implement Business Intelligence solution in the investment banking sector. The first part of this thesis is focused on general use of Business Intelligence in the banking sector and future developments in this area. Further it defines the objectives and reasons for the implementation of BI in the areas of investment banking. This part also explains basic terms used in investment banking which occur in the thesis and introduces a project in which the Business Intelligence is implemented. The second part focuses on determination of requirements and their analysis from which the data warehouse design is derived. Another part deals with the implementation of the designed data warehouse and ETL packages for loading the data. Further the tool in which the reports are created is described by exemplary implementation of the reports. In the last part of this thesis the implemented BI solution is evaluated by the users.

The Kimball methodology, often referred to as dimensional modelling, is well established in data warehousing and business intelligence as a highly successful means for turning data into information. Yet weaknesses exist in the Kimball approach that make it difficult to rapidly extend or interrelate dimensional models in complex business areas such as Health Care. This Thesis looks at the development of a methodology that will provide for the rapid extension and interrelation of Kimball dimensional models. This is achieved through the use of techniques similar to those employed in the semantic web. These techniques allow for rapid analysis and insight into highly variable data which previously was difficult to achieve. / Graduate

No description available.

The purpose of this qualitative research study was to identify difficulties some Swedish learners have with listening comprehension in the English 5 course in Upper Secondary Schools and to find methods of teaching such as scaffolding, schematic knowledge, preparation and peer cooperation that may prove successful. The methodology for this study was to test three groups of students representing three different programs of study at a high school in Sweden. These groups performed two different tasks to test theories about listening comprehension methods, and data was collected through test results and observations in the classroom, which were subsequently analyzed through discussion and comparison. The study showed that affective filters affected learning. However, scaffolding, schematic knowledge, preparation, and peer cooperation proved successful as reflected in higher all-around test scores. The researcher has attempted to identify the main aspects that make English listening comprehension difficult for Swedish learners, and to come up with ideas on how to break down the barriers to learning in order to promote improved listening comprehension in both live classrooms and in online learning situations.