In a learning community, conventional discussion forums are integral to webbased interventions in traditional classrooms as well as on-line learning environments. Despite popular belief that they are a great success in fostering deep and meaningful discussions and support active learning; research has found that there are millions of messages posted by users to express such an opinion, but it is hard to be directly delivered to all users. Finally there are millions of postings in databases across the country stored away and never reused. This thesis introduces a PhD student’s current research work. It proposes a distributed intelligent discussion forum system dedicated to supporting both students and teachers. The system is developed with the primary goal of reducing the number of problems associated with conventional discussion forum systems in web-based environments and improving the effectiveness of dialogue between students with each other and with teachers so that it can enhance each individual student’s ability to share and learn knowledge.

This thesis seeks to address current problems encountered when training classifiers within the framework of cascades of boosted ensembles (CoBE). At present, a signifi- cant challenge facing this framework are inordinate classifier training runtimes. In some cases, it can take days or weeks (Viola and Jones, 2004; Verschae et al., 2008) to train a classifier. The protracted training runtimes are an obstacle to the wider use of this framework (Brubaker et al., 2006). They also hinder the process of producing effective object detection applications and make the testing of new theories and algorithms, as well as verifications of others research, a considerable challenge (McCane and Novins, 2003). An additional shortcoming of the CoBE framework is its limited ability to train clas- sifiers incrementally. Presently, the most reliable method of integrating new dataset in- formation into an existing classifier, is to re-train a classifier from beginning using the combined new and old datasets. This process is inefficient. It lacks scalability and dis- cards valuable information learned in previous training. To deal with these challenges, this thesis extends on the research by Barczak et al. (2008), and presents alternative CoBE frameworks for training classifiers. The alterna- tive frameworks reduce training runtimes by an order of magnitude over common CoBE frameworks and introduce additional tractability to the process. They achieve this, while preserving the generalization ability of their classifiers. This research also introduces a new framework for incrementally training CoBE clas- sifiers and shows how this can be done without re-training classifiers from beginning. However, the incremental framework for CoBEs has some limitations. Although it is able to improve the positive detection rates of existing classifiers, currently it is unable to lower their false detection rates.

Dans le cadre de cette recherche, nous posons la problématique suivante : comment représente-t-on la découverte botanique en sciences naturelles ? Notre démarche sera sémiotique et particulièrement fondée sur la théorie greimassienne. Nous la lierons à une recherche sémantique qui s’appuiera notamment sur les travaux de François Rastier. Et parce que la représentation scientifique articule étroitement le texte linguistique au texte visuel, nous engagerons notre étude dans une perspective qui rendra compte du système visuel. Ce dernier sera interrogé à partir des écrits de Maria Giulia Dondero et Jacques Fontanillle, d’Anne Beyaert-Geslin ou encore de Jean-Marie Klinkenberg. A partir de ces fondements théoriques, cet objectif de recherche nous permettra de comprendre les mécanismes à l’œuvre lorsqu’il s’agit de représenter le vivant « non humain ». Le texte sera au cœur de notre recherche. Il supposera que nous l’abordions du point de vue de l’énoncé, mais aussi de l’énonciation. Pour envisager ce second volet, la théorie de l’énonciation d’Emile Benveniste sera convoquée. Parce que nous voulons mettre à l’épreuve la théorie, notre recherche intégrera une dimension empirique à partir d’un projet expérimental de lecture fondé sur la technique de l’eye-tracking (ou oculométrie). Celle-ci se montrera pertinente pour interroger la signification à partir de nos modes d’expression systématisés qui construisent le discours scientifique de la découverte depuis au moins deux siècles. A partir d’un corpus circonscrit issu des publications scientifiques du Muséum national d’histoire naturelle, nous définirons et expliquerons le parcours de la découverte et ses stratégies monstratives. Cela nous mènera à questionner nos modes de pensée. Pensons en effet que la médiation de nos représentations (le paraître) est le reflet d’une dimension ontologique (l’être). Elle fait de nous des instances connaissantes, conscientes et capables d’agir en conséquence. / As part of this research, we ask the following question: how do we represent the botanical discovery in natural sciences? Our approach will be semiotic and particularly based on Greimassian theory, which we will also link to a semantic analysis derived from the works of François Rastier. As scientific representation closely articulates linguistic text and visual text, this study will engage our object with a perspective that takes into account the visual system. The latter will be questioned in particular through the writings of Maria Giulia Dondero, Jacques Fontanillle, Anne Beyaert-Geslin and Jean-Marie Klinkenberg. With this theoretical foundation, our research objective will allow us to understand the mechanisms at work regarding the representation of non-human life-forms. Text will be at the heart of our research. He will suppose that we approach him from the point of view of the utterance, but also of the enunciation. This second part of our research will rely on the theory of the enunciation developped by Emile Benveniste. As our study aims to test the theory, therefore our research will integrate an empirical dimension based on an experimental reading project that uses eye-tracking. This will allow us to question meaning in the systematized modes of expression that have constructed scientific discourse in discovery for at least two centuries. Within a circumscribed corpus from the scientific publications of the National Museum of Natural History, we will define and explain the course of discovery and its monstrative strategies. This will lead us to question our ways of thinking. Let us think that the mediation of our representations (the appearance) is the reflection of an ontological dimension (being). It makes us knowledgeable, conscious and able to act accordingly.

Quel rôle est tenu par la fiction dans la relation qui s’établit entre des enfants - âgés entre 6 et 12 ans - et des monuments ? Financée par la région PACA, cette recherche s’intéresse aux formes prises par le monument dans sa « circulation triviale » (Jeanneret, 2008) et aux appropriations qu’en font les enfants. Partant du constat que la fiction nourrit les médiations proposées aux jeunes dans les monuments mais aussi que l’exposition médiatique de ces derniers est importante dans les fictions enfantines - notamment à travers le motif du château , trois niveaux de recherche ont été pensés. Le premier consiste à identifier, via une analyse sémiotique, les opérateurs de la fiction dans les médiations éducatives des monuments désignées explicitement comme telles - audioguides, livrets de visite, ateliers. La fiction mobilisée dans un tel contexte occupe des fonctions qui s’entrecroisent : didactique, ludique et expressive. Dans une démarche similaire, le second niveau propose d’analyser les processus de fabrication de monuments stéréotypés circulant massivement dans les fictions des industries culturelles - en prenant pour exemple, deux monographies : celle du château rêvé de Disney et de l’école Poudlard dans la saga Harry Potter. Le monument y agit principalement de façon cathartique, axiologique et diégétique. De fait, ces deux premiers niveaux mettent en exergue l’opérativité socio-culturelle de la fiction dans la relation proposée au monument. Mais que produisent ces opérateurs fictionnels ? Comment circulent-ils, se réinscrivent-ils et quelles valeurs confèrent-t-ils à la monumentalité ? Pour répondre à ces questions, le dernier niveau s’intéresse à leur réinscription dans les discours d’enfants sous une forme singulière de production de sens - appelés microliens. Ces derniers sont entendus comme des connexions infimes créées par analogie par des sujets entre des opérateurs de dispositifs actualisés dans des expériences sociales - la visite par exemple - et participant à l’élaboration d’un « être culturel » - la monumentalité. Les résultats montrent que la fiction permet des appropriations singulières du temps et de l’espace du monument et différentes formes d’immersion fictionnelle prises et de réflexivité de l’expérience vécue chez les enfants. Pour ces raisons, les expériences liées au monument seraient « polychrésiques » c’est-à-dire objets de constantes réappropriations et « prises(e) sans cesse dans un large spectre de logiques sociales différentes » (Jeanneret, op.cit. : 83). Le décloisonnement des domaines de la vie sociale que propose l’étude de la « trivialité », en tant que circulation des savoirs, montre que la naissance ou le développement d’une pratique pourrait se concevoir dans et avec un « ailleurs » imprégné de discours circulant comprenant des moments, des objets, des discours en dehors des monuments, autrement dit dans un espace ventilé.. / Which role is played by fiction in the relationship established between children - aged from 6 to 12 years old - and historical monuments? Funded by the PACA region, this research focuses on the forms taken by the historical monument in its « trivial circulation » (Jeanneret, 2008) and the appropriations that children do. Noting that fiction feeds mediations offered to young people in the monuments but also that the media exposure of these is important in children’s fictions - particularly through the pattern of the castle – three levels of search are thought. The first is to identify, through a semiotic analysis, operators of fiction in educational mediations of monuments explicitly designated as such – audio guides, visit booklets, workshops. Mobilized fiction in this context holds positions that intersect themselves: didactic, playful and expressive. In a similar move, the second level offers to analyse the manufacturing processes of stereotyped monuments massively flowing in the fictions of cultural industries - taking for example, two monographs: the dreamed castle of Disney and the Hogwarts School of Witchcraft and Wizardry in Harry Potter saga. The monument mainly acts in a cathartic, axiological and diegetic way. In fact, these two first levels highlight the socio-cultural operability of fiction in the proposed relation to the monument. But what do these fictional operators produce? How do they run, re-enroll themselves and what values do they give to “monumentality”? To answer these questions, the last level is interested in their re-enrollment in children’s speeches in a singular form of sense of production - called microlinks. These are understood as tiny connections created by analogy by subjects between updated devices operators in social experiences - the visit for example - and participating in the development of a « cultural being » - monumentality. The results show that fiction allows singular appropriations of time and space of the monument and different forms of fictional immersion taken and of reflexivity of the experience lived by children. For these reasons, the experiences linked to the monument would be « polychrésiques » which means objects of constant reappropriations and « taken constantly in a wide spectrum of different social logics » (Jeanneret, op.cit.: 83). The compartmentalisation of areas of social life that includes the study of the « triviality » as knowledge circulation, shows that the birth or the development of a practice can be conceived in and with an « elsewhere » imbued with running speech including moments, objects, speeches outside monuments, in other words in a ventilated space.

An open question in autism research is how to assess language abilities in this population. We investigated language development in monolingual and bilingual children with varying degrees of autism, ages 3 to 9, with the aim of better understanding vocabulary comprehension. Two different methodologies were used: the Receptive One-Word Picture Vocabulary Test (ROWPVT) and eye-tracker technique. We examined whether the eye-tracker could help in the assessment of these children because it does not require the child to point during the test. Four typically developing control children, 14 monolingual English children with moderate/mild autism, and 4 children (2 monolingual English, 2 bilingual Spanish/English) with severe autism were tested and the results of the ROWPVT test were compared to the eye-tracker results. Interestingly, bilingual children with severe autism had better results using eye-tracker than the traditional ROWPVT test. These results suggest that these children know more vocabulary than traditional test measures indicate.

No description available.

Parkinson ’s disease (PD) is a progressive neurological disorder that affects one’s movement. As the disease develops, individuals begin to present with symptoms that include but are not limited to bradykinesia, rigidity, tremors, and hypokinetic dysarthria. These symptoms affect a person’s entire body, including his/her voice. The Lee Silverman Voice Treatment (LSVT) program for treating individuals with PD is supported by over twenty-five years of research. It is considered a safe, non-invasive method to improve vocal loudness and speech clarity in individuals with PD. However, simply because LSVT is effective in its current state, it does not mean that the protocol is the most efficient or effective it can be. One potential shortcoming of LSVT is that it does not provide patients with much, if any, visual feedback. We hypothesized that visual feedback would enable the client to more easily produce a voice characterized by increased loudness and vocal quality. The results of the study do not fully support this hypothesis. There was less variability in the client’s performance within each session during the experimental weeks the patients performance over the course of the week improved, this pattern was not observed during the non-experimental weeks. Additionally, the participant expressed preference for treatment days when the visual feedback was used, finding it helpful in more effectively regulating the volume of his voice.

Objective: To determine the prevalence and odds ratios for cleft lip and/or palate (CL/P) among infants prenatally exposed to opioids with or without neonatal opioid withdrawal syndrome (NOWS). Design: This study represents an exploratory, retrospective cohort study design of newborn medical health records from 2011 to 2016. Setting: Records were drawn from a regional health system located in South Central Appalachia. Population and study sample: The original population yielded 3 cohorts of infants: (1) infants with opioid exposure (OE) but not requiring pharmacological intervention (OE; N = 168); (2) infants with NOWS requiring pharmacological intervention (N = 294); and (3) infants with no opioid exposure (NOE; N = 16 090), the primary comparison group. Main outcome: Infants in the NOWS and OE groups showed significantly increased prevalence and odds ratios for CL/P when compared to those in the NOE group. Results: Prevalence rates per 1000 live births for infants with OE (35.71) and infants with NOWS (6.80) were significantly higher than those for infants with NOE (1.37). Comparison of infants with OE to the NOE group revealed significantly increased odds for CL/P, isolated cleft palate (CP), cleft lip (CL), and cleft lip and palate (CLP) (27.05, 41.81, 19.26, 19.37, respectively; all Ps < .008). The odds ratios for infants with NOWS compared to the NOE group were significantly higher for CL/P and CP (5.00 and 10.98, respectively; Ps < .03) but not for CL and CLP. Conclusion: The results provide additional evidence that prenatal OE should be considered among the critical environmental risk factors that can contribute to CL/P.

The purpose of this study was to survey special education teachers and speech therapists in eleven school districts in Northeast Tennessee regarding their knowledge level (etiology and educational programming) of autism spectrum disorder (ASD). The primary focus of the study was to identify effective programs and methods used by special educators in this region, comparing them to the latest techniques and teaching methods prescribed by recent research. In addition, identified weaknesses were used to recommend future training and staff development to enable educators to provide the best possible programs for children with autism. Five hundred fifty-two surveys were disseminated to special education teachers and speech therapists in eleven school districts in Northeast Tennessee. Two hundred ninety-two professionals responded to the survey, resulting in a return rate of 52.9%. Educators were asked to respond to a total of 44 questions (28 true/false items and 16 multiple choice items). The multiple choice items were designed to obtain demographic information, job related characteristics, preparation and experience teaching students with ASD, and professional needs of special educators in this region. The 16 multiple choice items were categorized into knowledge of ASD etiology and ASD educational programming. The results of the study indicate that there were no marked deficits in special educators' knowledge levels (etiology and educational programming) of ASD. However, the scores on educational programming were consistently higher than scores on etiology. There is a need for further training because very few special educators have been trained in research based methods currently used with students diagnosed as ASD.

The speech of individuals with Parkinson's disease (PD) is typically characterized as lacking in proper prosody because of its monopitch and monoloud quality, in addition to its reduced intensity. These qualities make it difficult for others to understand speakers with PD. The purpose of the current study was to identify what individuals with PD would do vocally, if anything at all, to improve speech production following a simulated misunderstanding of what they had just said. The study evaluated the performance of 5 individuals with PD and compared their performance to 5 age- and sex-matched controls. Specifically, measures of vocal intensity (loudness), fundamental frequency (pitch), and utterance duration were made for repetitions of a ‘misheard’ phrase. In one experimental condition noise was presented through headphones to induce the Lombard effect. Both individuals with PD and controls used increased duration as a means of enhancing clarity in a repetition. Fundamental frequency (F0) and sound pressure level (SPL) were not consistently modified in repetitions for clarification. Under most speaking conditions, individuals with PD and controls had similar F0 and SPL. Individuals with PD, like the controls, responded to the presentation of masking noise by increasing their fundamental frequency and their intensity. Therefore, not all individuals with PD exhibit difficulty using prosody.