L'apprentissage statistique propose un vaste ensemble de techniques capables de construire des modèles prédictifs à partir d'observations passées. Ces techniques ont montré leurs capacités à traiter des volumétries importantes de données sur des problèmes réels. Cependant, de nouvelles applications génèrent de plus en plus de données qui sont seulement visibles sous la forme d'un flux et doivent être traitées séquentiellement. Parmi ces applications on citera : la gestion de réseaux de télécommunications, la modélisation des utilisateurs au sein d'un réseau social, le web mining. L'un des défis techniques est de concevoir des algorithmes permettant l'apprentissage avec les nouvelles contraintes imposées par les flux de données. Nous proposons d'abord ce problème en proposant de nouvelles techniques de résumé de flux de données dans le cadre de l'apprentissage supervisé. Notre méthode est constituée de deux niveaux. Le premier niveau utilise des techniques incrémentales de résumé en-ligne pour les flux qui prennent en compte les ressources mémoire et processeur et possèdent des garanties en termes d'erreur. Le second niveau utilise les résumés de faible taille, issus du premier niveau, pour construire le résumé final à l'aide d'une méthode supervisée performante hors-ligne. Ces résumés constituent un prétraitement qui nous permet de proposer de nouvelles versions du classifieur bayésien naïf et des arbres de décision fonctionnant en-ligne sur flux de données. Les flux de données peuvent ne pas être stationnaires mais comporter des changements de concept. Nous proposons aussi une nouvelle technique pour détecter ces changements et mettre à jour nos classifieurs.

This study aimed to investigate the differences of different teacher-student matches in science gender stereotype threat and classroom observations on teacher-student interactions of different gender match and interviews on science teachers. The study developed two standardized scales for junior high school students, Science Gender Stereotype Scale and Science Identification Scale. A total of 1,224 students from 44 classes out of 11 junior high schools in Kaohsiung City were selected through judgmental sampling to complete questionnaires in the study. The results indicated: (1) students of female-male group (female teacher and male student) had stronger science identification than students of female-female group (female teacher and female student); students of male-male group (male teacher and male student) had stronger science identification than students of male-female group (male teacher and female student); and female students taught by male teacher had the lowest science identification (2) there was no significant difference of students¡¦ gender stereotype belief in sciences in different teacher-student matches (3) students of female-male group (female teacher and male student) had stronger perception of gender stereotype than students of female-female group (female teacher and female student); students of male-male group (male teacher and male student) had stronger perception of gender stereotype than students of male-female group (male teacher and female student). The results of observations and interviews indicated: (1) male students were more initiative and therefore had more opportunities to interact with teachers (2) female students of upper-intermediate level were more susceptible to perceive gender stereotype threat in science learning (3) the difference in male and female teachers may be attributed to the influence of female role model (4) degree of difficulty was a crucial component stimulating students¡¦ gender stereotype threat.

Before entering school, children begin their science learning with their parents at home. This study proposes that parents' beliefs and actions regarding science shape their children's knowledge and skills that they then bring to school. Studying parents' beliefs about and practices with their children within the topic area of science provided insight into their influence in helping their children make sense of the world. Therefore, the purpose of this study aimed to investigate parents' beliefs about children's science learning and their actions in facilitating their children's science learning when they visited a science museum from socio-cultural perspectives. To investigate this, a qualitative case study examining nine Taiwanese parents of kindergarteners was conducted. The study was conducted in two parts. Data sources included field notes, parent interviews, and documents such as pictures of the equipment these parents bought for their children. First, through interviews with parents, their beliefs about their children's science learning were identified and examined. Four parts including parents' gendered science beliefs, parents' perceived importance of science learning, parents' beliefs about how science learning should proceed, and parents' beliefs about their engagement in science learning were found. Part two of the study examined how these nine parents' beliefs guided them in making decisions when they interacted with their children in a science museum through observations and follow-up interviews. In most cases, parents' beliefs appeared to be important resources for helping them find a proper way to interact with their children. Three issues including the person who took the lead at the family visits, the quantity of parents' intervention, and the scaffolding strategies these parents employed were found in their interactions with their children. Parents were aware of why they behaved in particular ways: because of their beliefs. Based on the findings, the researcher suggested that parents' beliefs were an important mechanism for influencing children's science learning. A seemingly simple behavior, such as letting children explore one object longer than others, might reflect what was recognized as important in their beliefs. Lastly, the implications for early childhood educators, parents of young children, and future research were provided. / text

Fantasies are defined as byproducts of human imagination and mental activities to internalize unusual external objective stimulus. In the literature, utilization of fantasy in educational settings promoted intellectual and emotional improvements. However, the research implications from these fantasy research studies are mostly limited to traditional game design and classroom teaching. There are two research purposes in this study. The first is to examine how different types of fantasy and student fantasy proneness influence science learning, factual information acquisition, and game engagement in a 3D educational game environment called “Alien Rescue.” To accomplish this purpose, this research investigated the effects of fantasy type and fantasy proneness on science learning, factual information of alien characters, and game engagement. The second purpose of this study is to investigate student’s perception of the varying types of fantasy. To accomplish the second purpose of the study, this research inquired how student identified each type of fantasy and related his or her past experience to the embodied characteristics in alien characters. The participants of the study were 103 students who used Alien Rescue in four classes as their science curriculum for 10 days. The students in two classes were assigned to a treatment group using models with portrayal fantasy and the students in two classes were assigned to the other treatment group using models with creative fantasy. Employing mixed methods, this study analyzed both quantitative and qualitative data such as surveys and student interviews. The results in the quantitative part of the study showed that portrayal fantasy was effective for science learning, alien information acquisition, and game engagement. Specifically, the students who used portrayal fantasy models showed higher improvement of science knowledge and scored better on both alien information acquisition and game engagement. High fantasy proneness group also showed better game engagement. The finding with qualitative data showed that the students pointed out eight elements in identifying 3D fantasy objects, and those elements were relevant to the design elements that the researcher included in the 3D modeling procedure. The students also showed a perception pattern that they understood 3D game characters based upon previous experience regardless of fantasy type. The findings suggested that portrayal fantasy was effective in enhancing content learning, factual information acquisition, and engagement in educational games because the familiarity of the fantasy elements makes the identification of the fantasy characters easier and faster. However, too deep involvement in fantasy resulted in ineffective and inefficient learning outcomes. The findings also suggested that eight components of 3D models were essential elements in identifying fantasy game characters by learners as well as designing the 3D characters by game designers. / text

Thesis (Ph.D.) - University of Glasgow, 2008. / Ph.D. thesis submitted to the Faculty of Education, Department of Educational Studies, University of Glasgow, 2008. Includes bibliographical references. Print version also available.

Thesis (Ph. D.)--Rutgers University, 2007. / "Graduate Program in Computer Science." Includes bibliographical references (p. 133-136).

Thesis (Ph. D.)--Ohio State University, 2005. / Title from first page of PDF file. Document formatted into pages; contains xv, 328 p.; also includes graphics (some col.). Includes bibliographical references (p. 238-249). Available online via OhioLINK's ETD Center

This dissertation investigated the relation between epistemic cognition-epistemic aims and source beliefs-and learning outcome in an Internet-based research context. Based on a framework of epistemic cognition (Chinn, Buckland, & Samarapungavan, 2011), a context-specific epistemic aims and source beliefs questionnaire (CEASBQ) was developed and administered to 354 students from college-level introductory chemistry courses. A series of multitrait-multimethod model comparisons provided evidence for construct convergent and discriminant validity for three epistemic aims-true beliefs, justified beliefs, explanatory connection, which were all distinguished from, yet correlated with, mastery goals. Students' epistemic aims were specific to the chemistry topics in research. Multidimensional scaling results indicated that students' source evaluation was based on two dimensions-professional expertise and first-hand knowledge, suggesting a multidimensional structure of source beliefs. Most importantly, online learning outcome was found to be significantly associated with two epistemic aims-justified beliefs and explanatory connection: The more students sought justifications in the online research, the lower they tended to score on the learning outcome measure, whereas the more students sought explanatory connections between information, the higher they scored on the outcome measure. There was a significant but small positive association between source beliefs and learning outcome. The influences of epistemic aims and source beliefs on learning outcome were found to be above and beyond the effects of a number of covariates, including prior knowledge and perceived ability with online sources. / Educational Psychology

<p>This thesis introduces the idea of an interactive web-based visualization tool to support inquiry-based science learning. The problem that occurs when the teachers and students are discussing the collected data is that they are lacking a tool to display such large quantities of data. It is often hard to fully understand such data. This education tool makes use of different visualization approaches in order to support students while getting insights from their collected data. In this thesis I proposed and implemented an interactive web-based visualization tool that was used at a prototype level during the educational activities. The requirements and user needs led the development of this prototype. Requirement elicitations have been done as a part of the research project conducted by CeLeKT.</p><p> </p><p>For the development of this tool, it was necessary for the input of the teachers and students in order to get an understanding of the requirements. The initial inquiry of the teachers and students show the necessity and usefulness of an interactive web-based visualization tool to support learning practices.</p>

Nous nous intéressons ici à une interface cerveau-machine (BCI, Brain Computer Interface) permettant de commander une prothèse par la pensée. Le rôle du BCI est de décoder à partir de signaux électroencéphalographiques (EEG) le mouvement désiré par le sujet. Le coeur du BCI est un algorithme de classification caractérisé par le choix des descripteurs des signaux et des règles de décision. L'objet de cette thèse est de développer un système BCI précis, capable d'améliorer ses performances en cours d'utilisation et de s'adapter à l'utilisateur sans nécessiter de multiples sessions d'apprentissage. Nous combinons deux moyens pour y parvenir. Le premier consiste à augmenter la précision du système de décision en recherchant des descripteurs pertinents vis à vis de l'objectif de classification. Le second est d'inclure un retour de l'utilisateur sur le système de décision : l'idée est d'estimer l'erreur du BCI à partir de potentiels cérébraux évoqués, reflétant l'état émotionnel du patient corrélé au succès ou à l'échec de la décision prise par le BCI, et de corriger le système de décision du BCI en conséquence. Les principales contributions de la thèse sont les suivantes : nous avons proposé une méthode d'optimisation de descripteurs à bases d'ondelettes pour des signaux EEG multivoies ; nous avons quantifié théoriquement l'amélioration des performances apportée par le détecteur ; un simulateur du système corrigé et bouclé a été développé pour observer le comportement du système global et comparer différentes stratégies de mise à jour de l'ensemble d'apprentissage ; le système complet a été implémenté et fonctionne en ligne dans des conditions réelles.