In dieser Arbeit zeige ich in einem groß angelegten Experiment die Auswirkungen adaptiver Elemente in einer Online-Lernplattform. Ich werde darauf eingehen, dass die derzeitige Forschung zu Online-Lernplattformen für den L1-Erwerb hauptsächlich deskriptiv ist und dass nur wenige adaptive Lernumgebungen in der Praxis verbreitet sind. In dieser Dissertation werde ich ein Konzept entwickeln, wie adaptives Lernen in L1-Online-Lernplattformen integriert werden kann, und analysieren, ob dies zu verbesserten Lernerfahrungen führt. Dabei konzentriere ich mich auf die Effektivität und Fairness von Vorhersagen und Interventionen sowie auf die geeignete Softwarearchitektur für den Einsatz in der Praxis. Zunächst werden verschiedene Vorhersagemodelle entwickelt, die besonders in Blended-Learning-Szenarien nützlich sind. Anschließend entwickle ich ein Architekturkonzept (adaptive learning as a service), um bestehende Lernplattformen mithilfe von Microservices in adaptive Lernplattformen umzuwandeln. Darauf aufbauend wird ein groß angelegtes online-kontrolliertes Experiment mit mehr als 11.000 Nutzer*innen und mehr als 950.000 eingereichten Rechtschreibaufgaben durchgeführt. In einer abschließenden Studie werden die Vorhersagemodelle auf ihren algorithmischen Bias hin untersucht. Außerdem teste ich verschiedene Techniken zur Verringerung von Bias. Diese Arbeit bietet eine ganzheitliche Sicht auf das adaptive Lernen beim Online-L1-Lernen. Durch die Untersuchung mehrerer Schlüsselaspekte (Vorhersagemodelle, Interventionen, Architektur und Fairness) ermöglicht die Arbeit Schlussfolgerungen sowohl für die Forschung als auch für die Praxis. / In this work I show in a large scale experiment the effect of adding adaptive elements to an online learning platform. I will discuss that the current research on online learning platforms in L1 acquisition is mainly descriptive and that only few adaptive learning environments are prevalent in practice. In this dissertation, I will develop a concept on how to integrate adaptive L1 online learning and analyse if it leads to improved learning experiences. I focus on the effectiveness and fairness of predictions and interventions as well as on the suitable software architecture for use in practice. First, I develop different prediction models, which are particularly useful in blended classroom scenarios. Subsequently, I develop an architectural concept (adaptive learning as a service) to transform existing learning platforms into adaptive learning platforms using microservices. Based on this, a large-scale online-controlled experiment with more than 11,000 users and more than 950,000 submitted spelling tasks is carried out. In the final study, the prediction models are examined for their algorithmic bias, by comparing different machine learning models, varying metrics of fairness, and multiple demographic categories. Furthermore, I test various bias mitigation techniques. The success of bias mitigation approaches depends on the demographic group and metric. However, in-process methods have proven to be particularly successful. This work provides a holistic view of adaptive learning in online L1 learning. By examining several key aspects (predictive models, interventions, architecture, and fairness), the work allows conclusions to be drawn for both research and practice.

Previous research suggests that the optimal time of day (TOD) for cognitive function for young adults occurs in the afternoon and evening times (Allen, et al. 2008; May, et al. 1993). The implication is college students may be more successful if they schedule classes and tests in the afternoon and evening times, but in asynchronous learning environments, “class” and tests take place at any TOD (or night) a student might choose. The problem is that there may be a disadvantage for students choosing to take tests at certain TOD. As educators, we need to be aware of potential barriers to student success and be prepared to offer guidance to students. This research study found a significant negative correlation between TOD and assessment scores on tests taken between 16:01 and 22:00 hours as measured in military time. While this study shows that academic performance on asynchronous assessments was high at 16:00 hours, student performance diminished significantly by 22:00 hours. When efforts were taken to mitigate the extraneous variables related to test complexity and individual academic achievement, the effect TOD had on assessment achievement during this time period was comparable to the effect of test complexity on that achievement. However, when analyzed using a small sub-set of the data neither GPA nor TOD could be used to predict student scores on tests taken between 16:01 and 22:00 hours. Finally, individual circadian arousal types (evening, morning and neutral) (Horne & Ostberg, 1976) and actual TOD students took tests were analyzed to determine if synchrony, the match between circadian arousal type and peak cognitive performance, existed. The synchrony effect could not be confirmed among morning type students taking this asynchronous online course, but evidence suggests that synchrony could have contributed to student success for evening types taking this asynchronous online courses. The implication of this study is that online instructors, instructional designers and students should consider TOD as a factor affecting achievement in asynchronous online courses. Results of this research are intended to propose further research into TOD effects in asynchronous online settings, and to offer guidance to online students as well as online instructors and instructional designers faced with setting deadlines and advising students on how to be successful when learning online.

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第23315号 / 情博第751号 / 新制||情||128(附属図書館) / 京都大学大学院情報学研究科社会情報学専攻 / (主査)教授 緒方 広明, 教授 黒田 知宏, 教授 楠見 孝 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Student affect can play a profoundly important role in students' post-school lives. Understanding students' affective states within online learning environments in particular has become an important matter of research, as digital tutoring systems have the potential to intervene at the moment that students are struggling and becoming frustrated, bored or disengaged. However, despite the importance of assessing students' affective states, there is no clear consensus about what emotions are most important to assess, nor how these emotions can be best measured. This dissertation investigates students’ self-reports of their emotions and causal attributions of those emotions collected while they are solving math problems within a mathematics tutoring system. These self-reports are collected in two conditions: through limited choice Likert response and through open response text boxes. The conditions are combined with students’ cognitive attributions to describe epistemic (neither purely affective nor purely cognitive) emotions in order to explain the relationship between observable student behaviors in the MathSpring.org tutoring system and student affect. These factors include beliefs, expectations, motivations, and perceptions of ability and control. A special emphasis of this dissertation is on analyzing the role of causal attributions for the events and appraisals of the learning environment, as possible causes of student behaviors, performance, and affect.

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2016-04-13T15:28:01Z No. of bitstreams: 1 João Luiz Cavalcante Ferreira_.pdf: 1672669 bytes, checksum: 80b5c6fbc873c9f858b230e78855dd55 (MD5) / Made available in DSpace on 2016-04-13T15:28:01Z (GMT). No. of bitstreams: 1 João Luiz Cavalcante Ferreira_.pdf: 1672669 bytes, checksum: 80b5c6fbc873c9f858b230e78855dd55 (MD5) Previous issue date: 2016-02-25 / Nenhuma / A Educação a Distância (EaD) no Brasil tem se consolidado com diversos estudantes optando por essa modalidade de ensino para ampliar suas formações e realização profissional, no entanto ela enfrenta alguns obstáculos, como a resistência de educandos e educadores, desafios organizacionais, custos de produção e a questão da reprovação ou retenção de alunos. Um dos principais diferenciais dos cursos EaD é a grande quantidade de dados gerados pelas interações no ambiente educacional, o que abre novas possibilidades para estudar e compreender estas interações. A Mineração de Dados educacionais (MDE) é uma área de pesquisa interdisciplinar que lida com o desenvolvimento de métodos para explorar dados originados no contexto educacional. A Learning Analytics (LA) é outra área de pesquisa emergente. Ela busca medir, coletar, analisar e relatar dados sobre estudantes. O desafio dos pesquisadores é desenvolver métodos capazes de prever o desempenho dos estudantes de modo a possibilitar a intervenção de professores e tutores visando resgatar o estudante antes que reprove. Esta dissertação propõe o MD-PREAD, um modelo para predição de grupos de risco de reprovação em um ambiente de Educação a Distância. A técnica de árvore de decisão foi utilizada para possibilitar um diferencial quanto à possibilidade de interpretação dos dados gerados pelo uso dos métodos de predição, pois outros métodos, tais como Redes Neurais Artificiais possuem como deficiência justamente a dificuldade de identificar as causas que levam aos resultados das predições. O modelo foi prototipado na ferramenta de mineração RapidMiner. Um experimento foi realizado no Instituto Federal de Educação, Ciência e Tecnologia do Amazonas, no programa Universidade Aberta do Brasil, no Curso de Filosofia da educação. Foram feitas coletas de dados históricos de 10 disciplinas de um grupo de 30 aprendizes em dois semestres consecutivos, 2014/2 e 2015/1, o total de alunos matriculados foi de 125, o total de interações levantadas foi de 41070, o cálculo de predição considerou as médias das avaliações de 30 aprendizes, os desvios padrões das interações e suas respectivas situações. Estes dados serviram para compor o conjunto de treinamento necessário para a definição da regra de classificação que teve como predominante a acurácia de 55% e a confiabilidade Kappa de 0,22. Foi realizado um segundo processo de validação, após o experimento, considerou-se os 125 alunos e o melhor classificador encontrado foi o J48 com a acurácia de 84,05%, precisão de 77,08% e recall de 50,23%. Concluiu-se que o MD-PREAD é uma ferramenta de auxílio no prognóstico de grupos de risco de reprovação, uma vez que possibilitou a geração e disponibilização semanal destes grupos a um sistema de recomendação educacional externo. / E-learning in Brazil has been established with many students opting for this type of education to expand their training and professional achievement, however it faces some obstacles, such as resistance from students and educators, organizational challenges, production costs and the question of failure or retention of students. One of the main advantages of e-learning courses is the large amount of data generated by the interactions in the educational environment, which opens up new possibilities to study and understand these interactions. Educational Data Mining (EDM) is an area of interdisciplinary research that deals with the development of methods to explore data that originates in the educational context. Learning Analytics (LA) is another area of emerging research. It seeks to measure, collect, analyze and report data on students. The challenge for researchers is to develop methods to predict the performance of students in order to allow the intervention of teachers and tutors aiming to retrieve the student before failing. This thesis proposes the MD-PREAD, a model for predicting failure of risk groups in a e-learning environment. The decision tree technique was used to enable a difference as to whether the interpretation of the data generated by the use of prediction methods, since other methods such as Artificial Neural Networks that has as disability difficulty in identifying precisely the causes that lead to predictions results. The model was prototyped in RapidMiner mining tool. An experiment was conducted at the Federal Institute of Education, Science and Technology of Amazonas, the Open University of Brazil program in course Philosophy of education. Historical data collection of 10 disciplines from a group of 30 apprentices were made in two consecutive semesters, 2014/2 and 2015/1, the total number of enrolled students was 125, the total raised interactions were 41070, the prediction calculation considered average of 30 apprentices ratings, the standard deviations of the interactions and their situations. These data served to compose the training set required for classification rule defining which had as predominant accuracy of 55% and Kappa reliability 0.22. A second validation process was carried out after the experiment. It was considered the total amount of 125 apprentices and the best classifier found was the J48 with the accuracy of 84.05%, 77.08% of classification precision and recall of 50.23%. It was concluded that the MD-PREAD is a support tool in the prognosis of failure risk groups, since it enabled the generation and weekly availability of these groups to a recommendation system.

In every learning or training environment, exercises are the basis for practical learning. Learners need to practice in order to acquire new abilities and perfect those gained previously. However, not every exercise is valid for every learner: learners require exercises that match their ability levels. Hence, difficulty of an exercise could be defined as the amount of effort that a learner requires to successfully complete the exercise (its learning cost). Too high difficulties tend to discourage learners and make them drop out, whereas too low difficulties are perceived as unchallenging, resulting in loss of interest. Correctly estimating difficulties is hard and error-prone problem that tends to be done manually using domain-expert knowledge. Underestimating or overestimating difficulty generates a problem for learners, increasing dropout rates in learning environments. This paper presents a novel approach to improve difficulty estimations by using Neuroevolution. The method is based on measuring the computational cost that Neuroevolution algorithms require to successfully complete a given exercise and establishing similarities with previously gathered information from learners. For specific experiments presented, a game called PLMan has been used. PLMan is a PacMan-like game in which users have to program the Artificial Intelligence of the main character using a Prolog knowledge base. Results show that there exists a correlation between students’ learning costs and those of Neuroevolution. This suggests that the approach is valid, and measured difficulty of Neuroevolution algorithms may be used as estimation for student's difficulty in the proposed environment.

Enrollment in full-time, virtual, K-12 schools is increasing while mathematics performance in these institutions is lacking compared to national averages. Scholarly literature lacks research studies using learning analytics to better predict student outcomes via student learning management system (LMS) interactions, specifically in the low performing area of middle school mathematics. The theoretical framework for this study was a combination of Hrastinski's theory of online learning as online participation and Moore's 3 types of interactions model of online student behavior. The purpose of this study was to address the current research gap in the full-time, K-12 eLearning field and determine whether 2 types of student LMS interactions could predict mathematics course performance. The research questions were developed to determine whether student clicks navigating course content page(s) or the number of times a student accessed resources predicted student performance in a full-time, virtual, mathematics course after student demographic variables were controlled for. This quantitative study used archived data from 238 seventh grade Math 7B students enrolled from January 8th-10th to May 22nd-25th in two Midwestern, virtual, K-12 schools. Hierarchical regressions were used to test the 2 research questions. Student clicks navigating the course content pages were found to predict student performance after the effects of student demographic covariates were controlled for. Similarly, the number of times a student accessed resources also predicted student performance. The findings from this study can be used to advise actionable changes in student support, build informative student activity dashboards, and predict student outcomes for a more insightful, data-driven, learning experience in the future.

Submitted by Automação e Estatística (sst@bczm.ufrn.br) on 2018-07-26T17:05:20Z No. of bitstreams: 1 IvanMaxFreireDeLacerda_TESE.pdf: 3236688 bytes, checksum: f8269c0c3bdbd5a0b84b45f9f4181c93 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-07-26T19:49:52Z (GMT) No. of bitstreams: 1 IvanMaxFreireDeLacerda_TESE.pdf: 3236688 bytes, checksum: f8269c0c3bdbd5a0b84b45f9f4181c93 (MD5) / Made available in DSpace on 2018-07-26T19:49:52Z (GMT). No. of bitstreams: 1 IvanMaxFreireDeLacerda_TESE.pdf: 3236688 bytes, checksum: f8269c0c3bdbd5a0b84b45f9f4181c93 (MD5) Previous issue date: 2018-03-02 / O crescente aumento dos dados registrados em cursos ofertados na modalidade a distância proporciona a utilização de métodos computacionais adaptados a pesquisa e agrupamento de dados educacionais, visando a descoberta de comportamentos de aprendizado. Essa área de pesquisa possibilita o desenvolvimento de ferramentas automatizadas de acompanhamento, predição e intervenção visando o aprimoramento dos índices educacionais. Em virtude disso, este trabalho propõe uma metodologia para a análise de programas de ensino a distância com base na tecnologia Learning Analytics, utilizando os dados de acesso dos alunos ao Ambiente Virtual de Aprendizagem (AVA), identificando os padrões sequenciais de uso mais frequentes e classificando-os de acordo com as categorias de aprendizagem autorregulada. Para a mineração sequencial de dados sequenciais os algoritmos SPAM e VGEN foram aplicados aos bancos de dados de duas instituições educacionais. Além do desenvolvimento da metodologia, como resultado desse processamento, uma grande incidência de um comportamento não previsto pela teoria da aprendizagem autorregulada foi identificado, e para classifica-lo foi criado um padrão chamado baixa participação. / The increasing of the data registered in courses offered in the distance modality boost the use of computational methods adapted to the research and the grouping of educational data, aiming to discover learning behaviors patterns. This research area allows the development of automated monitoring, prediction and intervention tools aiming at improving the educational indexes. As a result, this work proposes a methodology for analyzing distance learning programs based on the Learning Analytics technology, using the students’ access data to the Learning Management System (LMS), identifying the most frequent sequential patterns of use and classifying them as according to the self-regulated learning categories. For a sequential mining of sequential data the SPAM and VGEN algorithms were applied to the databases of two educational institutions. In addition to the development of the methodology, as a result of processing, a high incidence of behavior not predicted in the self-regulated learning theory was identified, and to classify it was created a pattern called low participation.

Submitted by Lucienne Costa (lucienneferreira@ufcg.edu.br) on 2018-05-03T20:37:40Z No. of bitstreams: 1 DOUGLAS AFONSO TENÓRIO DE MENEZES – TESE (PPGCC) 2017.pdf: 30688716 bytes, checksum: aeb667914da2d303938d7e10953623eb (MD5) / Made available in DSpace on 2018-05-03T20:37:40Z (GMT). No. of bitstreams: 1 DOUGLAS AFONSO TENÓRIO DE MENEZES – TESE (PPGCC) 2017.pdf: 30688716 bytes, checksum: aeb667914da2d303938d7e10953623eb (MD5) Previous issue date: 2017-07 / Capes / O uso de ambientes virtuais de aprendizagem está cada vez mais frequente, e muitas vezes os dados que são gerados nestes ambientes não são explorados adequadamente, o que dificulta a geração de indicadores da qualidade dos programas de formação. A qua- lidade do aprendizado em um ambiente virtual de aprendizagem é determinada por uma organização adequada do material e das formas de ensino. Para tal, deverá ser levado em consideração, entre outros fatores, o histórico de sucessos e insucessos de realizações anteriores das disciplinas a serem ofertadas e do perfil específico das novas turmas de alu- nos. O tratamento adequado destes dados pode evidenciar indicativos importantes sobre o desempenho de uma turma, tais como o nível de comprometimento e a motivação dos alunos, fatores que podem influenciar diretamente no processo de aprendizagem. Estes dadospodemserutilizadosporespecialistasaoseremexibidoscomoumresumoemforma de visualizações gráficas adequadas, possibilitando uma rápida interpretação e percepção de indicativos importantes dos cursos e seus alunos. As visualizações auxiliam na com- preensão e análise dos dados gerados, ampliam a cognição e facilitam a compreensão das informações apresentadas. Esta tese apresenta uma proposta para visualização de dados educacionais, com o objetivo de analisar como a visualização de dados pode ajudar o professor a identificar e adequar um Design Instrucional problemático, por meio de dados históricos, de cursos já realizados, assim como durante a realização de uma nova edição do curso, auxiliando-o na melhoria do DI de um curso/disciplina. / The use of virtual learning environments is becoming more frequent, and often the data that are generated in these environments are not properly exploited, which makes it difficult to generate quality indicators of training programs. The quality of learning in these settings is determined by an appropriate organization of the material and forms of instruction. In order to do so, it should be taken into account, among other factors, the history of successes and failures of previous achievements of the disciplines to be offered and the specific profile of a new class of students. Adequate treatment of these data can show important indicators about a class’s performance, such as the level of commitment and motivation of the students, factors that can directly influence the learning process. These data can be used by specialists to be displayed as a summary in the form of ade- quate graphical visualizations, allowing a fast interpretation and perception of important indications of a course and its students. It helps to understand a set of data, facilitate the analysis of the generated data, to increase the cognition and the understanding of the presented information. This thesis presents a proposal for the visualization of educatio- nal data, in which the objective is to analyze how the visualization of data can help the teacher to identify and adapt a problematic Instructional Design, through historical data of courses already carried out, and during the realization of a new edition of the course, assisting the teacher in improving the Instructional Design of a course / discipline.

<p>CAL (Computer Assisted Learning) programs are widespread today in schools and families due to the effectiveness of CAL programs in improving students’ learning and task performance. The flourishing of CAL programs in education has brought large amounts of students’ learning data including log data, performance data, mouse movement data, eye movement data, video data, etc. These data can present students’ learning or problem-solving processes and reflect underlying cognitive processes. These data are valuable resources for educators to comprehend students’ learning and difficulties. However, few data analysis methods can analyze and present CAL data for educators quickly and clearly. Traditional video analysis methods can be time-consuming. Current visualization analysis methods are limited to simple charts or visualizations of a single data type. In this dissertation, I propose a visual learning analytic approach to analyze and present students' problem-solving data from CAL programs. More specifically, a visualization system was developed to present students’ problem-solving data, including eye movement, mouse movement, and performance data, to help educational researchers understand student problem-solving processes and identify students’ problem-solving strategies and difficulties. An evaluation experiment was conducted to compare the visualization system with traditional video analysis methods. Seven educational researchers were recruited to diagnose students’ problem-solving patterns, strategies, and difficulties using either the visualization system or video. The diagnosis task loads and evaluators’ diagnosis processes were measured and the evaluators were interviewed. The results showed that analyzing student problem-solving tasks using the proposed visualization method was significantly quicker than using the video method. In addition, diagnosis using the visualization system can achieve results at least as reliable as the video analysis method. Evaluators’ preferences between the two methods are summarized and illustrated in the dissertation. Finally, the implications of the visual analytic approach in education and data visualization areas are discussed. </p>