Over the past several decades, research that explored various teaching-and-learning contexts has provided valuable insights into teacher-learner interactional practices in second language classrooms. Many of these practices focus on learners’ language accuracy by targeting the correct answer, a worthy but perhaps insufficient goal; an additional teacher responsibility is to encourage learners to build on their understanding by reasoning through that correct answer. This current study adds to previous research by examining how one experienced teacher and her adult ESL students in a community language program in the U.S. engage in a particular type of interactive, collaborative work on understanding that moves beyond what is correct to why it is correct, which I call “digging.” Based on a conversation analytic examination of 15 hours of video-recorded classroom interaction, the findings showcase two complementary types of teacher-led digging that are preceded by a critical “pre-digging” phase, during which the teacher redirects learners’ attention and constitutes a group that will work together as a collective. The first type of digging zooms in on one particular language issue which the teacher frames as a language challenge for the group and works collaboratively with the collective toward resolving it. The second type of digging, by contrast, zooms out from a specific language issue to a larger pattern in either the learners’ native languages or the target language, English. In both types of digging, exploratory talk and various scaffolding techniques are employed to promote participation and learner agency. The findings contribute to the literature on classroom interaction by specifying, in fine-grained detail, the how-to of these teacher interactional practices during whole group work on understanding which involves the intricate work of every gaze, every gesture, every posture shift, every utterance, and every second of silence. Such specifications also enrich teacher educators’ pedagogical content knowledge by providing them a common language to talk about, and illuminate the complexity of, teaching as they guide students to “see” such complexity.

Denna undersökning är en empirisk studie som söker att analysera vilka typer av frågor lärare ställer i sin undervisning och vilken kunskap dessa möjliggör för eleverna. Som grund för analysen ligger intervjuer med två lärare på mellanstadiet och observationer av dessas lektioner. Genom att använda ett sammandrag av intervjuerna har syftet varit att åskådliggöra lärarnas tankar kring frågor i undervisningen, och visa vad som ligger till grund för deras användning av dessa under de observerade lektionerna. De observerade lektionerna och analysen av dessa är undersökelsens huvudfokus. Genom analysen av observationerna gick det att se att frågor används i alla aspekter av lektionerna och att och att nära nog all kommunikation mellan lärare och elever försiggår med frågor som bas. Frågor används således inte enbart till att överföra kunskap , men också till att exempelvis motivera eller disciplinera elever. Något man kunde se i analysen var att alla frågetyperna som är definierade i undersökningen förekom under observationerna. Således erbjuds eleverna varierade former för kunskap, men kontrollerande och slutna frågor som gärna används till ren faktaförmedling dominerar. En annan intressant aspekt vid observationerna var att upptäcka att frågorna inte alltid var av den typ man skulle tro. Ibland fungerade exempelvis frågor som i formuleringarna framstod som öppna mera som slutna frågor, och ibland tvärt om.

The diploma thesis deals with the topic of board games. It is divided into three parts, the theoretical, practical and didactic part. In the theoretical part, the author describes iconography of board games and playing board games from the Middle Ages to the present time. The use of board games is interpreted on specific works of art in painting, sculpture and crafts. The practical part was focused on the creation of an art history educational board game, which develops knowledge in the form of a quiz with questions and answers. The game was redesigned as a digital / online version and a mobile application. Versions of the game were created in a unified author's design. In the didactic part, the author examines whether the game is a suitable teaching material for teachers, students and a group of respondents from the general public. It also includes a methodology for secondary school teachers on how to deal with play in teaching.

Question and answering sites are useful in sharing the knowledge by answering questions. It is a medium of sharing knowledge. Quora is the fastest emerging effective Q&A site, which is the best source of knowledge. Here you can ask a question, and get help in getting answers from people with firsthand experience, and blog about what you know. In this paper, we are investigating and identifying potential experts who are providing the best solutions to the questioner needs. We have considered several techniques in identifying user as an expert or non-expert. We have targeted the most followed topics in Quora and finally came up with five topics: Mathematics, Politics, Technology, Sports and Business. We then crawled the user profiles who are following these topics. Each topic dataset has many special features. Our research indicates that experts are quite different from normal users and tend to produce high quality answers to as many questions as possible to gain their reputation. After evaluation, we got a limited number of experts who have potential expertise in specific fields, achieving up to 97% accuracy and 0.987 AUC.

No description available.

Vision and Language are two major modalities that humans rely on to perceive the environment and understand the world. Recent advances in Artificial Intelligence (AI) facilitate the development of a variety of vision-language tasks derived from diverse multimodal interactions in daily life, such as image captioning, image-text matching, visual question answering (VQA), text-to-image generation, etc. Despite the remarkable performance, most previous state-of-the-art models are merely specialized for a single vision-language task, which lack generalizability across multiple tasks. Additionally, those specialized models sophisticate the algorithm designs and bring redundancy to model deployment when dealing with complex scenes. In this study, we investigate developing unified approaches capable of solving various vision-language interactions in a multi-task manner. We argue that unified multi-task methods could enjoy several potential advantages: (1) A unified framework for multiple tasks can reduce human efforts in designing different models for different tasks; (2) Reusing and sharing parameters across tasks can improve efficiency; (3) Some tasks may be complementary to other tasks so that multi-tasking can boost the performance; (4) They can deal with the complex tasks that need a joint collaborating of multiple basic tasks and enable new applications. In the first part of this thesis, we explore unified multi-task models with the goal of sharing and reusing as many parameters as possible between different tasks. We started with unifying many vision-language question-answering tasks, such as visual entailment, outside-knowledge VQA, and visual commonsense reasoning, in a simple iterative divide-and-conquer framework. Specifically, it iteratively decomposes the original text question into sub-question, solves each sub-question, and derives the answer to the original question, which can uniformly handle reasoning of various types and semantics levels within one framework. In the next work, we take one step further to unify tasks of image-to-text generation, text-to-image generation, vision-language understanding, and image-text matching all in one single large-scale Transformer-based model. The above two works demonstrate the feasibility, effectiveness and efficiency of sharing the parameters across different tasks in a single model. Nevertheless, they still need to switch between different tasks and can only conduct one task at a time. In the second part of this thesis, we introduce our efforts toward simultaneous multi-task models that can conduct multiple tasks at the same time with a single model. It has additional advantages: the model can learn to perform different tasks or combinations of multiple tasks automatically according to user queries; the joint interaction of tasks can enable new potential applications. We begin with compounding spatial understanding and semantic understanding in a single multimodal Transformer-based model. To enable models to understand and localize local regions, we proposed a hybrid region representation that seamlessly bridges regions with image and text. Coupled with a delicately collected training dataset, our model can perform joint spatial and semantic understanding at the same iteration, and empower a new application: spatial reasoning. Continuing the above project, we further introduce an effective module to encode the high-resolution images, and propose a pre-training method that aligns semantics and spatial understanding in high resolution. Besides, we also couple the Optical Character Recognition (OCR) capability together with spatial understanding in the model and study the techniques to improve the compatibility of various tasks.

[pt] Os medidores de pH existentes no mercado indicam valores confiáveis, dentro de uma incerteza declarada para cada modelo fabricado. Entretanto, normalmente, não é citada a especificação em termos do tempo necessário para alcançar estes valores, devido ao tempo de resposta do instrumento. Esta dissertação de mestrado tem como objetivo desenvolver uma metodologia teórico-experimental para avaliar o desempenho do transdutor em condições transientes, indicando alguns valores típicos para medidores comerciais, e sua resposta na medição de pH, a partir do modelo testado em uma reação experimental, sendo o mesmo simulado em outras reações químicas para diferentes velocidades de reação.Através deste modelo qualificado, foram avaliados os erros devidos às respostas permanentes e transientes, mostrando em que condições o medidor de pH pode ser usado, a partir da comparação entre valores de pH real e valores de pH registrados pelo medidor, levando em consideração a temperatura e o tempo de reação. / [en] The electrochemical analysers available in the market assert trustworthy values applicable in a specific uncertainty range, which varies depending on the unit s model. Nevertheless, rarely is given any information on the time necessary to reach these values, which depend on the answer of the instrument. This Master s dissertation aims to develop an experimental theoretical methodology to evaluate the performance of the transducer in transient conditions, and obtain some typical values for commercial analysers, and its answer in the pH measurements in simulated chemical reactions in various reaction velocities. Through the qualified model, the errors due to the permanent and transient answers were evaluated, establishing the conditions pH electronic units can be used, in a pre-set controlled error, from the comparison between real pH values and displayed pH values in the analysers, considering the temperature and the reaction time.

This diploma thesis deals with a question as a part of educational communication. The aim of the thesis is to analyze educational dialogue at primary school with an emphasis on the function, frequency and type of teachers' and pupils' questions. The first part of the thesis explores the theoretical background of function, frequency and typology of questions. It involves also the topic of the waiting time after asking a question, identifying the pupil who should answer, the quality of pupil's answer as well as teacher's response to it. The theoretical background is complemented by research based on analyses of three video records of lessons and interviews with teachers. One of the key findings of the research is that teacher's style of questioning is determined by his conception of teaching and the aims set.

Dans ces travaux de thèse nous proposons de tirer parti de la théorie des ensembles flous afin d'améliorer les interactions entre les systèmes de bases de données et les utilisateurs. Les mécanismes coopératifs visent à aider les utilisateurs à mieux interagir avec les SGBD. Ces mécanismes doivent faire preuve de robustesse : ils doivent toujours pouvoir proposer des réponses à l'utilisateur. Empty set (0,00 sec) est un exemple typique de réponse qu'il serait désirable d'éradiquer. Le caractère informatif des explications de réponses est parfois plus important que les réponses elles-mêmes : ce peut être le cas avec les réponses vides et pléthoriques par exemple, d'où l'intérêt de mécanismes coopératifs robustes, capables à la fois de contribuer à l'explication ainsi qu'à l'amélioration des résultats. Par ailleurs, l'utilisation de termes de la langue naturelle pour décrire les données permet de garantir l'interprétabilité des explications fournies. Permettre à l'utilisateur d'utiliser des mots de son propre vocabulaire contribue à la personnalisation des explications et améliore l'interprétabilité. Nous proposons de nous intéresser aux explications dans le contexte des réponses coopératives sous trois angles : 1) dans le cas d'un ensemble pléthorique de résultats ; 2) dans le contexte des systèmes de recommandation ; 3) dans le cas d'une recherche à partir d'exemples. Ces axes définissent des approches coopératives où l'intérêt des explications est de permettre à l'utilisateur de comprendre comment sont calculés les résultats proposés dans un effort de transparence. Le caractère informatif des explications apporte une valeur ajoutée aux résultats bruts, et forme une réponse coopérative. / In this thesis, we are interested in how we can leverage fuzzy logic to improve the interactions between relational database systems and humans. Cooperative answering techniques aim to help users harness the potential of DBMSs. These techniques are expected to be robust and always provide answer to users. Empty set (0,00 sec) is a typical example of answer that one may wish to never obtain. The informative nature of explanations is higher than that of actual answers in several cases, e.g. empty answer sets and plethoric answer sets, hence the interest of robust cooperative answering techniques capable of both explaining and improving an answer set. Using terms from natural language to describe data --- with labels from fuzzy vocabularies --- contributes to the interpretability of explanations. Offering to define and refine vocabulary terms increases the personalization experience and improves the interpretability by using the user's own words. We propose to investigate the use of explanations in a cooperative answering setting using three research axes: 1) in the presence of a plethoric set of answers; 2) in the context of recommendations; 3) in the context of a query/answering problem. These axes define cooperative techniques where the interest of explanations is to enable users to understand how results are computed in an effort of transparency. The informativeness of the explanations brings an added value to the direct results, and that in itself represents a cooperative answer.

The current thesis explores conditions for participation in interview interaction. Drawing on the ethnomethodological idea that knowledge is central to participation in social situations, it examines how interview participants navigate knowledge and competence claims and the institutional and moral implications of these claims. The data consists of, in total, 97 audio-recorded interviews conducted as part of a national Swedish evaluation of support interventions for children exposed to violence. In three studies, I use discursive psychology and conversation analysis to explicate how interview participants in interaction (1) contribute to and negotiate institutional constraints and (2) manage rights and responsibilities related to knowledge. The findings of study I and study II show that child interviewees actively cooperate with as well as resist the constraints of interview questions. However, the children’s opportunities for participation in this institutional context are limited by two factors: (1) recordability; that is, the focus on generating recordable responses and (2) problematic assumptions underpinning questions and the interpretation of interview answers. Apart from restricting children’s rights to formulate their experiences, these factors can lead interviewers to miss opportunities to gain important information. Also related to institutional constraints, study III shows how the ideal of model consistency is prioritized over service-user participation. Thus, the three studies show how different practices relevant to institutional agendas may hinder participation. Moreover, the findings contribute to an understanding of how issues of knowledge are managed in the interviews. Study II suggests the importance of the concept of believability to refer to people’s rights and responsibilities to draw conclusions about others’ thoughts. And the findings of study III demonstrate how, in evaluation interviews with social workers, children’s access to their own thoughts and feelings are based on a notion of predetermined participation; that is, constructed as contingent on wanting what the institutional setting offers. Thus, child service users’ low epistemic status, compared to the social workers, trumps their epistemic access to their own minds. These conclusions, about recordability, believability, and predetermined participation, are based on interaction with or about children. However, I argue that the findings relate to interviewees and service users in general. By demonstrating the structuring power of interactive practices, the thesis extends our understanding of conditions for participation in the institutional setting of social research interviews.