The benefits of observing a model when acquiring a new motor skill are well known, however, there is little research on the influence of viewing angle of the model. The purpose of the present experiment was to assess whether a looking-glass (face on) or subjective (facing away) viewing angle would result in different acquisition and retention levels when learning a complex Zumba dance sequence. Greater cognitive effort was expected during the looking-glass condition, consequently resulting in slower acquisition but greater physical performance scores and error recognition/identification. Thirty females were evenly divided into the looking-glass or subjective group and began with the pre-test phase to assess degrees of motivation, self-efficacy, and physical performance. Participants were then lead through six acquisition dances, within which they performed the to-be-learned sequence 18 times. An assessment of cognitive effort followed, then post-test performances and error recognition/identification scores were obtained to conclude the study. While both the looking-glass and subjective conditions demonstrated equal rates of acquisition (p>.05), the looking-glass group performed significantly fewer errors during the post-test (p<.05) and were significantly better at identifying errors when a video of the dance sequence was shown from the same viewing angle as the acquisition phase (p<.05). No differences were reported between the two conditions with respect to cognitive effort (p>.05). Based on the results of this study, the looking-glass viewing angle appears to result in better learning of a dance sequence, but cannot be explained by cognitive effort.

The benefits of observing a model when acquiring a new motor skill are well known, however, there is little research on the influence of viewing angle of the model. The purpose of the present experiment was to assess whether a looking-glass (face on) or subjective (facing away) viewing angle would result in different acquisition and retention levels when learning a complex Zumba dance sequence. Greater cognitive effort was expected during the looking-glass condition, consequently resulting in slower acquisition but greater physical performance scores and error recognition/identification. Thirty females were evenly divided into the looking-glass or subjective group and began with the pre-test phase to assess degrees of motivation, self-efficacy, and physical performance. Participants were then lead through six acquisition dances, within which they performed the to-be-learned sequence 18 times. An assessment of cognitive effort followed, then post-test performances and error recognition/identification scores were obtained to conclude the study. While both the looking-glass and subjective conditions demonstrated equal rates of acquisition (p>.05), the looking-glass group performed significantly fewer errors during the post-test (p<.05) and were significantly better at identifying errors when a video of the dance sequence was shown from the same viewing angle as the acquisition phase (p<.05). No differences were reported between the two conditions with respect to cognitive effort (p>.05). Based on the results of this study, the looking-glass viewing angle appears to result in better learning of a dance sequence, but cannot be explained by cognitive effort.

This thesis aims to build a system to tackle the task of diagnosing the grammatical errors in sentences written by learners of Chinese as a foreign language with the help of the CRF model (Conditional Random Field). The goal of this task is threefold:  1) identify if the sentence is correct or not, 2) identify the specific error types in the sentence, 3) find out the location of the identified errors. In this thesis, the task of Chinese grammatical error diagnosis is approached as a sequence tagging problem. The data and evaluation tool come from the previous shared tasks on Chinese Grammatical Error Diagnosis in 2016 and 2017. First, we use characters and POS tags as features to train the model and build the baseline system. We then notice that there are overlapping errors in the data. To solve this problem, we adopt three approaches: filtering out the problematic data, assigning encoding to characters with more than one label and building separate classifiers for each error type. We continue to increase the amount of training data and include syntactic features. The results show that both filtering out the problematic data and including syntactic features have a positive impact on the results. In addition, difference between domains of training data and test data can hurt performance to a large extent.

The benefits of observing a model when acquiring a new motor skill are well known, however, there is little research on the influence of viewing angle of the model. The purpose of the present experiment was to assess whether a looking-glass (face on) or subjective (facing away) viewing angle would result in different acquisition and retention levels when learning a complex Zumba dance sequence. Greater cognitive effort was expected during the looking-glass condition, consequently resulting in slower acquisition but greater physical performance scores and error recognition/identification. Thirty females were evenly divided into the looking-glass or subjective group and began with the pre-test phase to assess degrees of motivation, self-efficacy, and physical performance. Participants were then lead through six acquisition dances, within which they performed the to-be-learned sequence 18 times. An assessment of cognitive effort followed, then post-test performances and error recognition/identification scores were obtained to conclude the study. While both the looking-glass and subjective conditions demonstrated equal rates of acquisition (p>.05), the looking-glass group performed significantly fewer errors during the post-test (p<.05) and were significantly better at identifying errors when a video of the dance sequence was shown from the same viewing angle as the acquisition phase (p<.05). No differences were reported between the two conditions with respect to cognitive effort (p>.05). Based on the results of this study, the looking-glass viewing angle appears to result in better learning of a dance sequence, but cannot be explained by cognitive effort.

No description available.

Usability studies are important in today’s context. However, the increased security level of authentication systems is reducing the usability level. Thus, to provide secured but yet usable authentication systems is a challenge for researchers to solve till now. Learnability and human errors are influential factors of the usability of authentication systems. There are not many specific studies on the learnability and human errors concentrating on authentication systems. The authors’ aim of this study is to explore the human errors and the learnability situation of authentication systems to contribute to the development of more usable authentication systems. The authors investigated through observations and interviews to achieve the aim of this study. A minimalist portable test lab was developed in order to conduct the observation process in a controlled environment. At the end of the study, the authors showed the list of identified human errors and learnability issues, and provided recommendations, which the authors believe will help researchers to improve the overall usability of authentication systems. To achieve the aim of the study, the authors started with a systematic literature review to gain knowledge on the state of art. For the user study, a direct investigation, in form of observations and interviews was then applied to gather more data. The collected data was then analyzed and interpreted to identify and assess the human errors and the learnability issues. / This study addressed the usability experiences of users by exploring the human errors and the learnability situation of the authentication systems. Authors conducted a case study to explore the situation of human errors and learnability of authentication systems. Observation and interviews were adapted to gather data. Then analysis through SHERPA (to evaluate human errors) and Grossman et al. learnability metric (to evaluate learnability) had been conducted. First, the authors identified the human errors and learnability issues on the authentication systems from user’s perspective, from the gathered raw data. Then further analysis had been conducted on the summary of the data to identify the features of the authentication systems which are affecting the human errors and learnability issues. The authors then compared the two different categories of authentication systems, such as the 1-factor and the multi-factor authentication systems, from the gathered information through analysis. Finally, the authors argued the possible updates of the SHERPA’s human error metric and additional measurable learnability issues comparing to Grossman et al. learnability metrics. The studied authentication systems are not human errors free. The authors identified eight human errors associated with the studied authentication systems and three features of the authentication systems which are influencing the human errors. These errors occurred while the participants in this study took too long time locating the login menu or button or selecting the correct login method, and eventually took too long time to login. Errors also occurred when the participants failed to operate the code generating devices, or failed to retrieve information from errors messages or supporting documents, and/or eventually failed to login. As these human errors are identifiable and predictable through the SHERPA, they can be solved as well. The authors also found the studied authentication systems have learnability issues and identified nine learnability issues associated with them. These issues were identified when very few users could complete the task optimally, or completed without any help from the documentation. Issues were also identified while analyzing the participants’ task completion time after reviewing documentations, operations on code generating devices, and average errors while performing the task. These learnability issues were identified through Grossman et al. learnability metric, and the authors believe more study on the identified learnability issues can improve the learnability of the authentication systems. Overall, the authors believe more studies should be conducted on the identified human errors and learnability issues to improve the overall human errors and learnability situation of the studied authentication systems at presence. Moreover, these issues also should be taken into consideration while developing future authentication systems. The authors believe, in future, the outcome of this study will also help researchers to propose more usable, but yet secured authentication systems for future growth. Finally, authors proposed some potential research ares, which they believe will have important contribution to the current knowledge. In this study, the authors used the SHERPA to identify the human errors. Though the SHERPA (and its metrics) is arguably one of the best methods to evaluate human errors, the authors believe there are scopes of improvements in the SHERPA’s metrics. Human’s perception and knowledge is getting changed, and to meet the challenge, the SHERPA’s human error metrics can be updated as well. Grossman et al. learnability metrics had been used in this study to identify learnability issues. The authors believe improving the current and adding new metrics may identify more learnability issues. Evaluation of learnability issues may have improved if researchers could have agreed upon a single learnability definition. The authors believe more studies should be conducted on the definition of learnability in order to achieve more acceptable definition of the learnability for further research. Finally, more studies should be conducted on the remedial strategies of the identified human errors, and improvement on the identified learnability issues, which the authors believe will help researchers to propose more usable, but yet secured authentication systems for the future growth. / 30/1, Shideshwari Lane, Shantinagar, Ramna, Dhaka, Bangladesh, Post Code 1217. Contact: +88017130 16973

No description available.

Les travaux de recherche présentés dans ce mémoire concernent des contributions à l'identification paramétrique de modèles à temps continu. La première contribution est le développement d'une méthode à erreur de sortie appliquée à des modèles linéaires, en boucle ouverte et en boucle fermée. Les algorithmes sont présentés pour des modèles à temps continu, en utilisant une approche hors ligne ou récursive. La méthode est étendue à l'identification de systèmes linéaires comprenant un retard pur. La méthode développée est appliquée à différents systèmes et comparée aux méthodes d'identification existantes. La deuxième contribution est le développement d'une nouvelle approche d'identification de systèmes à boucles imbriquées. Cette approche est développée pour l'identification de systèmes électromécaniques. Elle se base sur l'utilisation d'un modèle d'identification paramétrique générique d'entraînements électromécaniques en boucle fermée, sur la connaissance du profil des lois de mouvement appliquées appelées excitations, et sur l'analyse temporelle de signaux internes et leurs corrélations avec les paramètres à identifier. L'approche est développée dans le cadre de l'identification d'entraînements à courant continu et synchrone. L'application de cette approche est effectuée au travers de simulations et de tests expérimentaux. Les résultats sont comparés à des méthodes d'identification classiques. / The research works presented in this thesis are about contributions in continuous time model parametric identication. The rst work is the development of an output error method applied on linear models, in open and closed loop. The algorithms are presented for continuous time models, using in-line or oine approaches. The method is extended to the case of the linear systems containing pure time delay. The developed method is applied to several systems and compared to the best existing methods. The second contribution is the development of a new identication approach for cascaded loop systems. This approach is developed for identifying electromechanical systems. It is based on the use of a generic parametric model of electromechanical drives in closed loop, on the knowledge of the movement laws applied and called excitations, and on the analyse of the time internal signals and their correlations with the parameters to identify. This approach is developed for identifying direct current and synchronous drives. The approach is applied with simulations and experimental tests. The obtained results are compared to best identifying known methods.

State estimation plays an important role in modern power systems. The errors in the telemetered measurements and the connectivity information of the network will greatly contaminate the estimated system state. This dissertation provides solutions to suppress the influences of these errors. A two-stage state estimation algorithm has been utilized in topology error identification in the past decade. Chapter II discusses the implementation of this algorithm. A concise substation model is defined for this purpose. A friendly user interface that incorporates the two-stage algorithm into the conventional state estimator is developed. The performances of the two-stage state estimation algorithms rely on accurate determination of suspect substations. A comprehensive identification procedure is described in chapter III. In order to evaluate the proposed procedure, a topology error library is created. Several identification methods are comparatively tested using this library. A remote measurement calibration method is presented in chapter IV. The un-calibrated quantities can be related to the true values by the characteristic functions. The conventional state estimation algorithm is modified to include the parameters of these functions. Hence they can be estimated along with the system state variables and used to calibrate the measurements. The measurements taken at different time instants are utilized to minimize the influence of the random errors. A method for auto tuning of measurement weights in state estimation is described in chapter V. Two alternative ways to estimate the measurement random error variances are discussed. They are both tested on simulation data generated based on IEEE systems. Their performances are compared. A comprehensive solution, which contains an initialization process and a recursively updating process, is presented. Chapter VI investigates the errors introduced in the positive sequence state estimation due to the usual assumptions of having fully balanced bus loads/generations and continuously transposed transmission lines. Several tests are conducted using different assumptions regarding the availability of single and multi-phase measurements. It is demonstrated that incomplete metering of three-phase system quantities may lead to significant errors in the positive sequence state estimates for certain cases. A novel sequence domain three-phase state estimation algorithm is proposed to solve this problem.

Made available in DSpace on 2016-06-02T19:06:12Z (GMT). No. of bitstreams: 1 5932.pdf: 1110060 bytes, checksum: fe08b552e37f04451248c376cfc4454f (MD5) Previous issue date: 2014-04-10 / Universidade Federal de Minas Gerais / The project described in this document focusses on the post-editing of automatically translated texts. Machine Translation (MT) is the task of translating texts in natural language performed by a computer and it is part of the Natural Language Processing (NLP) research field, linked to the Artificial Intelligence (AI) area. Researches in MT using different approaches, such as linguistics and statistics, have advanced greatly since its beginning in the 1950 s. Nonetheless, the automatically translated texts, except when used to provide a basic understanding of a text, still need to go through post-editing to become well written in the target language. At present, the most common form of post-editing is that executed by human translators, whether they are professional translators or the users of the MT system themselves. Manual post-editing is more accurate but it is cost and time demanding and can be prohibitive when too many changes have to be made. As an attempt to advance in the state-of-the-art in MT research, mainly regarding Brazilian Portuguese, this research has as its goal verifying the effectiveness of using an Automated Post-Editing (APE) system in translations from English to Portuguese. By using a training corpus containing reference translations (good translations produced by humans) and translations produced by a phrase-based statistical MT system, machine learning techniques were applied for the APE creation. The resulting APE system is able to: (i) automatically identify MT errors and (ii) automatically correct MT errors by using previous error identification or not. The evaluation of the APE effectiveness was made through the usage of the automatic evaluation metrics BLEU and NIST, calculated for post-edited and not post-edited sentences. There was also manual verification of the sentences. Despite the limited results that were achieved due to the small size of our training corpus, we can conclude that the resulting APE improves MT quality from English to Portuguese. / O projeto de mestrado descrito neste documento tem como foco a pós-edição de textos traduzidos automaticamente. Tradução Automática (TA) é a tarefa de traduzir textos em língua natural desempenhada por um computador e faz parte da linha de pesquisa de Processamento de Línguas Naturais (PLN), vinculada à área de Inteligência Artificial (IA). As pesquisas em TA, utilizando desde abordagens linguísticas até modelos estatísticos, têm avançado muito desde seu início na década de 1950. Entretanto, os textos traduzidos automaticamente, exceto quando utilizados apenas para um entendimento geral do assunto, ainda precisam passar por pós-edição para que se tornem bem escritos na língua alvo. Atualmente, a forma mais comum de pós-edição é a executada por tradutores humanos, sejam eles profissionais ou os próprios usuários dos sistemas de TA. A pós-edição manual é mais precisa, mas traz custo e demanda tempo, especialmente quando envolve muitas alterações. Como uma tentativa para avançar o estado da arte das pesquisas em TA, principalmente envolvendo o português do Brasil, esta pesquisa visa verificar a efetividade do uso de um sistema de pós-edição automática (Automated Post-Editing ou APE) na tradução do inglês para o português. Utilizando um corpus de treinamento contendo traduções de referência (boas traduções produzidas por humanos) e traduções geradas por um sistema de TA estatística baseada em frases, técnicas de aprendizado de máquina foram aplicadas para o desenvolvimento do APE. O sistema de APE desenvolvido: (i) identifica automaticamente os erros de TA e (ii) realiza a correção automática da tradução com ou sem a identificação prévia dos erros. A avaliação foi realizada usando tanto medidas automáticas BLEU e NIST, calculadas para as sentenças sem e com a pós-edição; como analise manual. Apesar de resultados limitados pelo pequeno tamanho do corpus de treinamento, foi possível concluir que o APE desenvolvido melhora a qualidade da TA de inglês para português.