The object of the study is machine learning methods used to assess student performance in online courses. The goal of the final qualifying work is to develop a tool that uses machine learning methods to assess student performance in online educational courses. Methods or methodology for carrying out work: analytical research, software implementation. The subject of the work is to determine the basic concepts in assessing the performance of students in online courses. This paper presents the development of a machine learning-based system for assessing student performance in online courses. The main goal is to create an efficient, scalable and reliable tool that automates the process of predicting academic performance and provides useful information for students, teachers and the educational platform. The research includes comprehensive domain analysis, task analysis, model training and economic evaluation. Key contributions include the analysis and implementation of a CATboost-based machine learning model, which achieved strong performance with an ROC-AUC of 0.88 and a balanced accuracy of 0.78. The study also details the process of data preparation, tuning of model hyperparameters, and integration of the system with existing learning management systems (LMS). Cost-benefit and product performance analyzes demonstrate significant long-term benefits, highlighting high return on investment and improved operational efficiency. The system has a positive impact on various stakeholders by providing timely feedback and personalized support to students, reducing administrative burden on faculty, and increasing institutional reputation and revenue. Overall, this work highlights the potential of machine learning to predict student performance, improve learning outcomes, and create more effective educational environments. The developed system represents a valuable tool for educational institutions seeking to expand their online learning offerings and better support their students and staff. / Объектом исследования является методы машинного обучения применяемые для оценки успеваемости студентов на онлайн курсах. Целью выпускной квалификационной работы является разработка инструмента, который использует методы машинного обучения для оценки успеваемости студентов на образовательных онлайн курсах. Методы или методология проведения работы: аналитическое исследование, программная реализация. Предметом работы является определение основных концепций в оценке успеваемости обучающихся на онлайн курсах В данной работе представлена разработка системы на основе машинного обучения для оценки успеваемости студентов на онлайн-курсах. Основная цель — создать эффективный, масштабируемый и надежный инструмент, который автоматизирует процесс прогнозирования успеваемости и предоставляет полезную информацию для студентов, преподавателей и образовательной платформы. Исследование включает в себя комплексный анализ предметной области, анализ задач, обучение модели и экономическую оценку. Ключевой вклад включает анализ и внедрение модели машинного обучения на основе CATboost, которая достигла высоких показателей производительности с показателем ROC-AUC 0,88 и сбалансированной точностью 0,78. В исследовании также подробно описан процесс подготовки данных, настройка гиперпараметров модели и интеграция системы с существующими системами управления обучением (LMS). Анализ экономической эффективности и эффективности продукта демонстрирует значительные долгосрочные преимущества, подчеркивая высокую отдачу от инвестиций и повышение операционной эффективности. Система положительно влияет на различные заинтересованные стороны, обеспечивая своевременную обратную связь и персонализированную поддержку для студентов, снижая административную нагрузку на преподавателей, а также повышая репутацию и доходы учреждения. В целом, эта работа подчеркивает потенциал машинного обучения в прогнозировании успеваемости студентов, улучшении результатов обучения и создании более эффективной образовательной среды. Разработанная система представляет собой ценный инструмент для образовательных учреждений, стремящихся расширить свои предложения онлайн-обучения и улучшить поддержку своих студентов и сотрудников.

Huvudsyftet med studien har varit att utvärdera hur en fem timmar lång kommunikationsutbildning riktad till sjukgymnaster på habilitering påverkar sjukgymnasternas kommunikativa stil i interaktion med barn/ungdomar som på grund av cerebral pares saknar tal. Utbildningen utformades som en inlärningsmodell i åtta delmoment (åttastegsmodellen). I utbildningen och studien ingick sex habiliteringssjukgymnaster som interagerade med var sitt barn/ungdom som till följd av cerebral pares kommunicerar med alternativt och kompletterande kommunikation (AKK) i form av kommunikationskartor. Interaktionen videofilmades vid tre separata tillfällen: en inspelning före utbildning, en inom två veckor efter utbildningen samt en sista filmning tre månader efter avslutad utbildning. Analys av videomaterialet gjordes med hjälp av en kombination av kvantitativ och kvalitativ analysmetod. Analysformuläret KOMMUNIKATIV användes för att analysera sjukgymnasternas kommunikation och kompletterades med en mer detaljerad kvalitativ samtalsanalys (Conversation Analysis, CA) där interaktionen mellan individerna analyserades. Utöver detta besvarade sjukgymnasterna en enkät. Utbildningens utformning baserades på tidigare forskning i form av av miljömodifierande strategier, responsiv kommunikationsstil samt AKK-modell i form av att pekprata. Resultatet av KOMMUNIKATIV påvisade en statistiskt signifikant förändring av sjukgymnasternas kommunikativa beteende efter avslutad utbildning. Det fanns även en fortsatt mätbar, men inte statistiskt signifikant, förändring mellan andra och tredje mättillfället. Sjukgymnasterna möjliggjorde, efter utbildning, att barnen/ungdomen i större utsträckning kunde kommunicera med sin kommunikationskarta. De pekpratade i större utsträckning och var mer lyhörda för barnets/ungdomens kommunikation. Endast en av tio kommunikativa förmågor som analyseras i KOMMUNIKATIV försämrades: sjukgymnastens förmåga att förtydliga sig. Samtalsanalysen bekräftade den kvantitativa analysen på flera sätt och visade att barnet/ungdomen efter utbildning anpassade sig till sjukgymnasternas förändrade beteende och uttryckte sig i längre fraser. Interaktionsmönstret ändrades från att före utbildning vara mer styrt av sjukgymnasten till att barnet/ungdomen efter utbildning kunde införa nya ämnen och delta i reparerande sekvenser. Samtalsanalysen visade även hur sjukgymnasterna ändrade sin användning av en engagerad röstkvalitet. Före kursen användes den i flera olika kontexter, inklusive problematiska kommunikativa kontexter där den snarare förvärrade problemen. Efter kursen varierade sjukgymnasterna sina strategier för att lösa olika typer av problematiska situationer, och undvek därmed i högre grad kommunikativa problem. Åttastegsmodellen som provades i genomförandet av kursen var framgångsrik på två olika sätt. Dels medgav den att varje deltagare fick en individuell målsättning baserad på en förmätning, och dels medförde modellen ett aktivt lärande vilket bidrog till att befästa den nya kunskapen. En slutsats av utvärderingen är att en interventionsutbildning enligt åttastegsmodellen med fördel kan användas för att lära ut kommunikativa strategier till habiliteringspersonal. / The main purpose of this study was to examine how a five hour communication course given to physiotherapists working at a habilitation center changes their communication when interacting with non-speaking children and teenagers with cerebral palsy. An instructional model for teaching learning strategies in eight steps was tested. Six physiotherapists took part of a five hour communication course on three occasions. The course contained receptive communication, environmental arrangements and aided language stimulation as suggested by previous research. Interaction between physiotherapist and child/teenager communicating with a communication board was videotaped just before the course, within two weeks after the course and also three month later. The course was analyzed using a combination of quantitative and qualitative methods. KOMMUNIKATIV is a quantitative method measuring ten communication abilities of the physiotherapists. Conversation Analysis (CA) was used as a detailed qualitative complementary analysis to KOMMUNIKATIV in order to examine the interaction between physiotherapist and child/teenager. Results from KOMMUNIKATIV showed a statistically significant change within two weeks after the course. Continued changes were measured between the second and the third point of measure but that change was not statistically significant. These results strongly indicate that the physiotherapists adopted a more receptive communication style after the course. They made it possible for the child/teenager to communicate with the communication board and they increased the number of AAC-modeling. One aspect, the physiotherapists’ ability to simplify the communication, declined from the first to the second and third point of measure. The child´s/teenager´s communication adapted in some ways to the changed communicational behavior of the physiotherapists. After the course, the child/teenager used longer phrases and started to make repairs. The detailed interaction analysis verified the change in communicative behavior in several ways. One significant change was the way the physiotherapists used an engaged voice quality to encourage the child/teenager to respond to requests for action. Before the course this voice quality was used in many different contexts, including communicatively problematic contexts, where it increased the problems. After the course the physiotherapists used a more varied set of strategies to solve problematic situations, thereby avoiding communicative problems. The instructional model for teaching communicative strategies in eight steps that was used in the study was successful in two ways. Firstly, the model provides the possibility to set individual goals for each member based on preassessments. Secondly, this model enables active learning which seem to consolidate the new ability. A conclusion from the evaluation of the eight step model is that it can be used for teaching communicative strategies to professionals.

1 ABSTRACT The thesis discusses possibilities of application of the concept of lifelong education in Basic Art Schools. The goal of the thesis is to propose a basic model of adult education which could be applied in Basic Art Schools. It focuses on the analysis of the literature on lifelong learning, lifewide learning, further education, and also on the analysis of strategic documents of the European Union and the Czech Republic in the field of lifelong learning and further education. Based on the analyses of the documents and external environment, a model of adult education in Basic Art Schools is proposed, the applicability of which is verified by a research investigation. A survey research examines the existence of education of diverse target groups of adults (both laymen and experts) in Basic Art Schools in Prague, the ratio of adult to younger students and the employed methods of their education. Structured interviews among five headteachers of Basic Art Schools in Prague clarify specific details of the process of educating adults in Basic Art Schools and provide more observations on the topic. The thesis contributes to the advancement of the concept of lifelong learning in the sphere of Basic Art Schools. Also, it brings forward possibilities of use of vacancies in Basic Art Schools for adult...

Des écrits scientifiques mettent en évidence les retombées de l’humanisation et de la déshumanisation des soins pour les patients. Les formateurs et les chercheurs mettent donc à l’essai de nombreuses stratégies pédagogiques prometteuses qui visent à promouvoir l’humanisation des soins. Or, les connaissances portant sur le processus d’apprentissage de la compétence « agir avec humanisme » demeurent peu développées. Cette thèse doctorale présente une étude phénoménologique qui visait à élaborer un modèle cognitif de l’apprentissage (MCA) de la compétence « agir avec humanisme », à partir de l’expérience vécue d’apprentissage d’étudiantes infirmières et de diplômées d’un programme de baccalauréat en sciences infirmières basé sur une approche par compétences, et à identifier les indicateurs de développement s’y rattachant. L’étude a été orientée par une perspective conceptuelle intégrant les sciences infirmières, soit le Modèle humaniste des soins infirmiers – UdeM (MHSI-UdeM) développé par Cara et al. (2016), et les sciences de l’éducation, soit la conception du développement des compétences de Tardif (2006). La philosophie humaniste, à la base du MHSI-UdeM, a été approfondie et analysée à la lumière de cinq conceptions infirmières. Une analyse du MHSI-UdeM selon la méthode de Chinn et Kramer (2018) a également été réalisée. La phénoménologie interprétative de Benner (1994), ancrée dans les référents philosophiques d’Heidegger (1927/2008), a été privilégiée. Les participantes (n = 26), des étudiantes et des infirmières, ont été recrutées dans six groupes en fonction de leur progression universitaire et professionnelle. Des entrevues individuelles ont été conduites puis les récits recueillis ont été transcrits et analysés selon une opérationnalisation de la perspective phénoménologique de Benner (1994). Le MCA de la compétence « agir avec humanisme » a été décliné en cinq niveaux de développement suivants : 1) conscientisation à l’humanisation des soins, 2) appropriation d’habiletés de communication investies dans une approche humaniste, 3) conciliation d’un idéal de pratique humaniste aux responsabilités de l’infirmière clinicienne, 4) intégration d’une approche humaniste dans l’ensemble des soins et 5) maîtrise d’une approche humaniste dans l’environnement de travail. D’autres résultats ont montré qu’un développement de la compétence « agir avec humanisme » commençait avant les études en sciences infirmières et qu’il était différent d’une étudiante à l’autre. Les résultats ont aussi mis en évidence des leviers et des contraintes au développement de la compétence « agir avec humanisme », résidant à la fois dans les institutions d’enseignement et les établissements de santé. Il y avait notamment la présence de modèles de rôle humanistes et d’environnements qui promeuvent l’humanisation des soins, ainsi que la surcharge de travail qui l’entrave. Les participantes ont aussi énoncé des recommandations visant à optimiser le développement de la compétence « agir avec humanisme » qui se sont avérées fortement cohérentes avec les leviers et les contraintes identifiées. Des retombées pour les cinq domaines de la pratique infirmière découlent de cette étude et offrent certaines pistes qui pourraient favoriser le développement de la compétence « agir avec humanisme ». L’une d’elles vise à harmoniser l’entrée dans la profession infirmière après la formation, une période charnière où le développement de la compétence « agir avec humanisme » s’est avéré être le plus vulnérable. / Several papers illustrate the outcomes of humanization and dehumanization of care for patients. Educators and researchers are testing many promising educational strategies aimed at promoting humanization of care. However, little is known about the development of the “humanistic caring” competency. This thesis presents a phenomenological study aimed at elaborating a cognitive learning model (CLM) of the “humanistic caring” competency, from the lived experiences of nursing students and graduates of a competency-based program, and to identify related developmental indicators. A conceptual perspective combining nursing, the Humanistic Model of Nursing Care – UdeM (HMNC-UdeM) developed by Cara et al. (2016), and education, a conception about competency development (Tardif, 2006), oriented the study. The humanistic philosophy at the roots of the HMNC-UdeM was scrutinized and analyzed in the context of five conceptual models. An analysis of the HMNC-UdeM, based on Chinn and Kramer’s method (2018), was also carried out. Benner’s (1994) interpretive phenomenology, grounded in the philosophical foundations of Heidegger (1927/2008), was used as the study methodology. The participants (n = 26), students and nurses, were recruited into six groups based on their progression in the education program and their clinical experience. Individual interviews were conducted, and the stories collected were transcribed and analyzed according to an operationalization of Benner’s (1994) phenomenological perspective. The CLM of the “humanistic caring” competency comprised five developmental stages: 1) conscientization to humanization of care, 2) assimilation of communication skills integrated in a humanistic approach, 3) reconciliation of an ideal of humanistic practice with the responsibilities of a nurse, 4) integration of a humanistic approach in nursing care, and 5) mastery of a humanistic approach in the work environment. Other results showed that a development of "humanistic caring" began before studying nursing and that it was different from one student to another. The results also highlighted facilitators and constraints to the development of "humanistic caring" residing in both educational institutions and healthcare facilities. These included the presence of humanistic role models and environments that promote humanization of care, and the work overload, that impedes it. The participants also suggested recommendations aimed at optimizing the development of "humanistic caring", and these were highly consistent with the aforementioned facilitators and constraints. Implications for the five domains of practice emerge from this study and offer some avenues that could promote the development of "humanistic caring". One of them is to facilitate the entry into the nursing profession after graduation, a turning point where the development of "humanistic caring" has proven to be the most vulnerable.

A major challenge in training machine learning models is the gathering and labeling of a sufficiently large training data set. A common solution is the use of synthetically generated data set to expand or replace a real data set. This paper examines the performance of a machine learning model trained on synthetic data set versus the same model trained on real data. This approach was applied to the problem of character recognition using a machine learning model that implements convolutional neural networks. A synthetic data set of 1’240’000 images and two real data sets, Char74k and ICDAR 2003, were used. The result was that the model trained on the synthetic data set achieved an accuracy that was about 50% better than the accuracy of the same model trained on the real data set. / Vid utvecklandet av maskininlärningsmodeller kan avsaknaden av ett tillräckligt stort dataset för träning utgöra ett problem. En vanlig lösning är att använda syntetiskt genererad data för att antingen utöka eller helt ersätta ett dataset med verklig data. Denna uppsats undersöker prestationen av en maskininlärningsmodell tränad på syntetisk data jämfört med samma modell tränad på verklig data. Detta applicerades på problemet att använda ett konvolutionärt neuralt nätverk för att tyda tecken i bilder från ”naturliga” miljöer. Ett syntetiskt dataset bestående av 1’240’000 samt två stycken dataset med tecken från bilder, Char74K och ICDAR2003, användes. Resultatet visar att en modell tränad på det syntetiska datasetet presterade ca 50% bättre än samma modell tränad på Char74K.