Second language learners face many challenges when learning a new language. To determine which challenges learners needed additional support in overcoming, we conducted a needs assessment of the Russian language program at the University of Saskatchewan and found that their students needed the most help with speaking in Russian. As a result, we designed an Intelligent Tutoring System (ITS) to help students learn how to pronounce Russian properly. We hoped to alleviate some of the challenges that learners face when learning to pronounce words in a second language by building an ITS that uses a Historic Open Learner Model (HOLM) to encourage learner reflection and to help maintain learner motivation. We designed, built, and performed a formative evaluation of a system, called ProTutor, using beginner learners of Russian as a second language at the University of Saskatchewan. This evaluation showed that learners have a positive perception of HOLMs and of the system as a whole. However, ProTutor needs further evaluation in order to determine its effectiveness as a learning aide.

Second language learners face many challenges when learning a new language. To determine which challenges learners needed additional support in overcoming, we conducted a needs assessment of the Russian language program at the University of Saskatchewan and found that their students needed the most help with speaking in Russian. As a result, we designed an Intelligent Tutoring System (ITS) to help students learn how to pronounce Russian properly. We hoped to alleviate some of the challenges that learners face when learning to pronounce words in a second language by building an ITS that uses a Historic Open Learner Model (HOLM) to encourage learner reflection and to help maintain learner motivation. We designed, built, and performed a formative evaluation of a system, called ProTutor, using beginner learners of Russian as a second language at the University of Saskatchewan. This evaluation showed that learners have a positive perception of HOLMs and of the system as a whole. However, ProTutor needs further evaluation in order to determine its effectiveness as a learning aide.

This thesis describes an Intelligent Tutoring System developed to be used in distance education at Master&rsquo / s level. The system was designed and implemented to help teachers to generate course material for online tutoring and to help the students to navigate through the course material according to their knowledge level. The system integrates many new technologies and provides individualized learning for students which is one of the most efficient ways for learning. How well the student has learned the course material is tested immediately after each smallest learning unit by end-of-section tests and the knowledge level of the student is derived from the answers given to these tests. This knowledge level is used to build a user model. This thesis describes how this user model is used for navigational support for students while studying on the course material.

Artificiell Intelligens har på senare tid exploderat i både kommersiell och privat användning, vilket förändrar hur vi arbetar, studerar och får tillgång till information. Vanliga typer av AI-tjänster är chatbots och språkmodeller. ChatGPT är en språkmodell som har blivit populär sedan tjänsten lanserades i november 2022. Med hjälp av ChatGPT kan användare exempelvis generera programmeringskod, få svar på frågor, undersöka olika ämnen och översätta mellan språk. Introduktionen av ChatGPT har skapat en debatt om hur den ska användas i lärandemiljöer. Idag har vi inte tillräcklig kunskap om hur man använder AI i utbildning på ett effektivt sätt, vilket syns i tidigare studier som visar att det finns en rädsla kring AI och utbildning. Däremot finns det stor potential i att använda språkmodeller som ChatGPT i utbildning eftersom det kan fungera som en handledare för varje elev på ett sätt som en mänsklig handledare inte kan tack vare den ständiga tillgängligheten av tjänsten. Forskningsfrågan för uppsatsen är: Hur använder studenter AI-tekniken ChatGPT som ett verktyg för självständigt lärande? En ytterligare underfråga formulerades med avsikten att utforska problematiska aspekter av att använda ChatGPT i utbildning: Vilka möjligheter och risker upplever studenter med ChatGPT som ett verktyg för självständigt lärande? Studien genomfördes med hjälp av en kartläggningsstrategi, med kvalitativa semistrukturerade intervjuer som datainsamlingsmetod. Nio intervjuer genomfördes med studenter som har använt ChatGPT flera gånger för sina studier. För att analysera intervjuerna användes en tematisk metod. Resultaten av studien visar några sätt som studenter använder ChatGPT på för självständigt lärande och inkluderar att använda språkmodellen för inspiration och idéer, be om förklaringar för att få en översikt över ett ämne, generera programmeringskod samt sammanfattning och omformulering av text. En allmän användning var att behandla ChatGPT som en digital handledare. Några av riskerna och nackdelarna med att använda ChatGPT som ett verktyg för självständigt lärande är felaktig information som produceras av språkmodellen, rädsla för plagiering och en oro för att bli alltför beroende av verktyget, vilket kan leda till en minskning av lärandet. Studien begränsas av det faktum att endast ett litet antal intervjuer kunde genomföras, vilket förhindrar generaliseringar från de resulterande uppgifterna. Resultatens överförbarhet kan bedömas genom att jämföra dem med framtida forskning inom området. Studien anses inte ha några negativa etiska eller samhälleliga konsekvenser, eftersom etiska aspekter har beaktats i studiens metodik. Detta uppnås genom att säkerställa fullständig anonymitet och frivilligt deltagande. Resultaten av studien kan användas för att integrera ChatGPT mer effektivt i inlärningsmiljöer, vilket gynnar både elever och lärare. / Artificial intelligence has recently exploded in both commercial and private usage which is changing the way we work, study and access information. Some common types of AI-services are chatbots and language models. ChatGPT is a language model that has become immensely popular since the service was launched in November 2022. Using ChatGPT, users can for example generate programming code, get answers to questions, research different topics and translate between languages. The introduction of ChatGPT has created a debate on how it should be used in learning environments. Today we do not have sufficient knowledge on how to use AI in education in an efficient manner, which is seen in earlier studies that show that there is a fear surrounding AI and education. On the other hand, there is great potential in using language models like ChatGPT in education since it can act like a tutor for each student in a way that a human tutor can’t thanks to the constant availability of service.The research question of the thesis is: How do students use the AI-technology ChatGPT as a tool for independent learning? An additional sub-question was formulated with the intention to explore problematic aspects of using ChatGPT in education: What potentials and risks do students experience when using ChatGPT as a tool for independent learning? The study is conducted using a survey study strategy, with qualitative semi-structured interviews as the data collection method. Nine interviews were performed, with students who have used ChatGPT several times for their studies. To analyze the interviews a thematic approach was used. The findings of the study show that some ways that students use ChatGPT for independent learning include using the language model for inspiration and ideas, asking for explanations to get an overview of a topic, generating programming code and summarization and reformulation of text. A general usage was to treat ChatGPT as a digital tutor. Some of the risks and cons with using ChatGPT as a tool for independent learning are found to be inaccurate information produced by the language model, fear of plagiarism and a worry of becoming too dependent on the tool, which can lead to a decrease in learning. The study is limited by the fact that only a small number of interviews could be conducted, which prevents drawing any generalizations from the resulting data. The transferability of the results can be assessed by comparing them with future research in the field. The study is considered to have no negative ethical or societal consequences, as ethical aspects have been taken into account in the study’s methodology. This is achieved by ensuring complete anonymity and voluntary participation. The results of the study can be utilized to incorporate ChatGPT more effectively in learning environments, benefiting both students and teachers.

This study focusses on artificial intelligence (AI) in education, AIEd, and on how AI language models, such as ChatGPT, impact teachers' ability to develop the pupils' skills and knowledge, specifically at an upper secondary school. Three language teachers at a Swedish upper secondary school have been interviewed with the purpose of unveil new insights within the purpose of this piece of research which is to contribute with an understanding of how to develop pupils' skills and knowledge through successfully integrating artificial intelligence in education (AIEd) in an upper secondary school. After defining the problem analysis and discussion for this thesis my conclusion is that there are several recent studies performed for K-12 schools but not conducted enough research for specifically upper secondary schools in relation to AI, since the introduction of ChatGPT and other large language models (LLMs). Therefore, the ambition with this thesis is to shrink the knowledge-gap in answering the research question and fullfilling the purpose of this thesis. The analysis in this thesis is a semantic thematic analysis and to facilitate reading the empirics and analysis parts every theme is included in the empirics so the connection to that latter analysis is displayed in the empirics and in this way constitutes a bridging between the empirics to the latter analysis. I hope you find the study worthwhile reading and if you are a teacher, it will bring some support in the choices you are facing in the era of LLMs in an upper secondary school.

Este trabalho propõe um agente pedagógico animado que possui o objetivo de fornecer suporte emocional ao aluno: motivando-o e encorajando-o, fazendo-o acreditar em suas próprias habilidades e promovendo um estado de espírito positivo no aluno, que é melhor para o seu aprendizado. Este suporte cuidadoso do agente, suas táticas afetivas, é expresso através de comportamentos emotivos e mensagens de encorajamento do personagem animado. Devido à tendência social humana de antropomorfizar software, nós acreditamos que um agente de software pode realizar esse papel afetivo. Para escolher as táticas afetivas adequadas, o agente deve conhecer as emoções do aluno. O agente proposto infere as seguintes emoções do aluno: alegria/tristeza, satisfação/frustração, raiva/gratidão e vergonha a partir do comportamento observável do aluno, isto é, as ações do aluno na interface do sistema educacional. A inferência das emoções é fundamentada psicologicamente na teoria cognitiva das emoções. Mais especificamente, nós usamos o modelo OCC o qual é baseado na abordagem cognitivista das emoções e é possível de ser implementado computacionalmente. Devido a natureza dinâmica da informação sobre o estado afetivo do aluno, nós adotamos uma abordagem BDI para implementar o modelo afetivo do usuário e o diagnóstico afetivo. Além disso, em nosso trabalho nós nos beneficiamos da capacidade de raciocínio do BDI para o agente deduzir o appraisal do aluno, que lhe permite inferir as emoções do aluno. Como um caso de estudo, o agente proposto é implementado como o Agente Mediador de MACES: um ambiente para ensino colaborativo à distância modelado com uma arquitetura multiagente e baseado psicologicamente na abordagem Sociocultural de Vygotsky. / This work proposes an animated pedagogical agent that has the role of providing emotional support to the student: motivating and encouraging him, making him believe in his self-ability, and promoting a positive mood in him, which fosters learning. This careful support of the agent, its affective tactics, is expressed through emotional behaviour and encouragement messages of the lifelike character. Due to human social tendency of anthropomorphising software, we believe that a software agent can accomplish this affective role. In order to choose the adequate affective tactics, the agent should also know the student’s emotions. The proposed agent recognises the student’s emotions: joy/distress, satisfaction/disappointment, anger/gratitude, and shame, from the student’s observable behaviour, i. e. his actions in the interface of the educational system. The inference of emotions is psychologically grounded on the cognitive theory of emotions. More specifically, we use the OCC model which is based on the cognitive approach of emotion and can be computationally implemented. Due to the dynamic nature of the student’s affective information, we adopted a BDI approach to implement the affective user model and the affective diagnosis. Besides, in our work we profit from the reasoning capacity of the BDI approach in order for the agent to deduce the student’s appraisal, which allows it to infer the student’s emotions. As a case study, the proposed agent is implemented as the Mediating Agent of MACES: an educational collaborative environment modelled as a multi-agent system and pedagogically based on the sociocultural theory of Vygotsky.

Este trabalho propõe um agente pedagógico animado que possui o objetivo de fornecer suporte emocional ao aluno: motivando-o e encorajando-o, fazendo-o acreditar em suas próprias habilidades e promovendo um estado de espírito positivo no aluno, que é melhor para o seu aprendizado. Este suporte cuidadoso do agente, suas táticas afetivas, é expresso através de comportamentos emotivos e mensagens de encorajamento do personagem animado. Devido à tendência social humana de antropomorfizar software, nós acreditamos que um agente de software pode realizar esse papel afetivo. Para escolher as táticas afetivas adequadas, o agente deve conhecer as emoções do aluno. O agente proposto infere as seguintes emoções do aluno: alegria/tristeza, satisfação/frustração, raiva/gratidão e vergonha a partir do comportamento observável do aluno, isto é, as ações do aluno na interface do sistema educacional. A inferência das emoções é fundamentada psicologicamente na teoria cognitiva das emoções. Mais especificamente, nós usamos o modelo OCC o qual é baseado na abordagem cognitivista das emoções e é possível de ser implementado computacionalmente. Devido a natureza dinâmica da informação sobre o estado afetivo do aluno, nós adotamos uma abordagem BDI para implementar o modelo afetivo do usuário e o diagnóstico afetivo. Além disso, em nosso trabalho nós nos beneficiamos da capacidade de raciocínio do BDI para o agente deduzir o appraisal do aluno, que lhe permite inferir as emoções do aluno. Como um caso de estudo, o agente proposto é implementado como o Agente Mediador de MACES: um ambiente para ensino colaborativo à distância modelado com uma arquitetura multiagente e baseado psicologicamente na abordagem Sociocultural de Vygotsky. / This work proposes an animated pedagogical agent that has the role of providing emotional support to the student: motivating and encouraging him, making him believe in his self-ability, and promoting a positive mood in him, which fosters learning. This careful support of the agent, its affective tactics, is expressed through emotional behaviour and encouragement messages of the lifelike character. Due to human social tendency of anthropomorphising software, we believe that a software agent can accomplish this affective role. In order to choose the adequate affective tactics, the agent should also know the student’s emotions. The proposed agent recognises the student’s emotions: joy/distress, satisfaction/disappointment, anger/gratitude, and shame, from the student’s observable behaviour, i. e. his actions in the interface of the educational system. The inference of emotions is psychologically grounded on the cognitive theory of emotions. More specifically, we use the OCC model which is based on the cognitive approach of emotion and can be computationally implemented. Due to the dynamic nature of the student’s affective information, we adopted a BDI approach to implement the affective user model and the affective diagnosis. Besides, in our work we profit from the reasoning capacity of the BDI approach in order for the agent to deduce the student’s appraisal, which allows it to infer the student’s emotions. As a case study, the proposed agent is implemented as the Mediating Agent of MACES: an educational collaborative environment modelled as a multi-agent system and pedagogically based on the sociocultural theory of Vygotsky.

Este trabalho propõe um agente pedagógico animado que possui o objetivo de fornecer suporte emocional ao aluno: motivando-o e encorajando-o, fazendo-o acreditar em suas próprias habilidades e promovendo um estado de espírito positivo no aluno, que é melhor para o seu aprendizado. Este suporte cuidadoso do agente, suas táticas afetivas, é expresso através de comportamentos emotivos e mensagens de encorajamento do personagem animado. Devido à tendência social humana de antropomorfizar software, nós acreditamos que um agente de software pode realizar esse papel afetivo. Para escolher as táticas afetivas adequadas, o agente deve conhecer as emoções do aluno. O agente proposto infere as seguintes emoções do aluno: alegria/tristeza, satisfação/frustração, raiva/gratidão e vergonha a partir do comportamento observável do aluno, isto é, as ações do aluno na interface do sistema educacional. A inferência das emoções é fundamentada psicologicamente na teoria cognitiva das emoções. Mais especificamente, nós usamos o modelo OCC o qual é baseado na abordagem cognitivista das emoções e é possível de ser implementado computacionalmente. Devido a natureza dinâmica da informação sobre o estado afetivo do aluno, nós adotamos uma abordagem BDI para implementar o modelo afetivo do usuário e o diagnóstico afetivo. Além disso, em nosso trabalho nós nos beneficiamos da capacidade de raciocínio do BDI para o agente deduzir o appraisal do aluno, que lhe permite inferir as emoções do aluno. Como um caso de estudo, o agente proposto é implementado como o Agente Mediador de MACES: um ambiente para ensino colaborativo à distância modelado com uma arquitetura multiagente e baseado psicologicamente na abordagem Sociocultural de Vygotsky. / This work proposes an animated pedagogical agent that has the role of providing emotional support to the student: motivating and encouraging him, making him believe in his self-ability, and promoting a positive mood in him, which fosters learning. This careful support of the agent, its affective tactics, is expressed through emotional behaviour and encouragement messages of the lifelike character. Due to human social tendency of anthropomorphising software, we believe that a software agent can accomplish this affective role. In order to choose the adequate affective tactics, the agent should also know the student’s emotions. The proposed agent recognises the student’s emotions: joy/distress, satisfaction/disappointment, anger/gratitude, and shame, from the student’s observable behaviour, i. e. his actions in the interface of the educational system. The inference of emotions is psychologically grounded on the cognitive theory of emotions. More specifically, we use the OCC model which is based on the cognitive approach of emotion and can be computationally implemented. Due to the dynamic nature of the student’s affective information, we adopted a BDI approach to implement the affective user model and the affective diagnosis. Besides, in our work we profit from the reasoning capacity of the BDI approach in order for the agent to deduce the student’s appraisal, which allows it to infer the student’s emotions. As a case study, the proposed agent is implemented as the Mediating Agent of MACES: an educational collaborative environment modelled as a multi-agent system and pedagogically based on the sociocultural theory of Vygotsky.

Inwieweit KI das neuartige universitäre Lehren und Lernen wirksam begleiten kann, wird im BMBF-Verbundprojekt 'tech4comp: Personalisierte Kompetenzentwicklung durch skalierbare Mentoringprozesse' untersucht. Gemeinsam beforscht man soziotechnische Artefakte für personalisiertes digital-gestütztes Mentoring für Studierende. Hierzu werden u.a. Rahmenbedingungen und (soziale) Kontextfaktoren erforscht, um die Implementierung von KI in der Hochschulbildung zu unterstützen. Es wird davon ausgegangen, dass unabhängig von der Art der Technologie und vom pandemischen Kontext, insbesondere die Akzeptanz und Bereitschaft der beteiligten Stakeholder zum erfolgreichen Einsatz intelligenter Bildungstechnologien beiträgt. Das ZQA/KfBH der TU Dresden widmet sich unter der Leitung von Dr. Cathleen M. Stützer im Forschungsprojekt der Elaboration von Handlungsfeldern, die sich aus einer soziotechnischen Beforschung von KI in der Hochschulbildung ergeben. Fallstudien hierzu stellen sich u. a. Fragen zu Gelingensbedingungen und Wirksamkeit digitaler Hochschulbildung, um (prospektiv) eine erfolgreiche Implementierung KI-gestützter adaptiver Mentoringsysteme mit evidenten Forschungsberichten zu unterstützen.:Vorwort & Danksagung Abbildungsverzeichnis Tabellenverzeichnis Abkürzungsverzeichnis 1. Einleitung 2. Methodik 3. Ergebnisse 4. Implikationen 4.1 Einflussfaktoren und Gelingensbedingungen der KI-Akzeptanz 4.2 Handlungsempfehlungen 5. Zusammenfassung und Fazit 6. Limitationen 7. Literaturverzeichnis Anhang / The extent to which AI can effectively accompany new types of university teaching and learning is being investigated in the BMBF joint project 'tech4comp: Personalised competence development through scalable mentoring processes'. Together, they are researching socio-technical artefacts for personalised digitally-supported mentoring for students. For this purpose, framework conditions and (social) contextual factors, among others, are being researched in order to support the implementation of AI in higher education. It is assumed that regardless of the type of technology and the pandemic context, the acceptance and willingness of the stakeholders involved in particular contributes to the successful use of intelligent educational technologies. Under the direction of Dr. Cathleen M. Stützer, the ZQA/KfBH at TU Dresden is dedicated to the elaboration of fields of action resulting from socio-technical research on AI in higher education. Case studies on this topic address questions such as the conditions for success and the effectiveness of digital higher education in order to (prospectively) support the successful implementation of AI-supported adaptive mentoring systems with evident research reports.:Vorwort & Danksagung Abbildungsverzeichnis Tabellenverzeichnis Abkürzungsverzeichnis 1. Einleitung 2. Methodik 3. Ergebnisse 4. Implikationen 4.1 Einflussfaktoren und Gelingensbedingungen der KI-Akzeptanz 4.2 Handlungsempfehlungen 5. Zusammenfassung und Fazit 6. Limitationen 7. Literaturverzeichnis Anhang

Dynamiska AI-system tar får allt mer utrymme i skolans värld. Det kommer påverka utbildning, undervisning och inlärning och inte minst lärares och studenters roller. Syftet med denna studie är därför att undersöka hur lärares och elevers roller, enligt de trender som framkommer i konferenserna AIED 2021 och LAK21, påverkas av AI-system i skolan. Materialet i denna studie utgörs av konferensartiklarna som presenterades under de två konferenser 2021 som iscensattes av International Society of Artificial Intelligence in Education (IAED) och Society for Learning Analytics (SoLAR). I sin forskning fokuserar båda forskningssällskapen användandet av AI i skolan. Båda ser sig som ledande inom sina fält. Av sammanlagt 187 har 130 uppfyllt de ställda kriterierna för att inkluderas i denna litteraturöversikt. Artiklarna har kategoriserats utifrån vilka AI-system som använts och deras påverkan på lärar- respektive studentrollen. Resultatet som framkommer visa på en mer passiv lärarroll och en mer aktiv studentroll. Förklaringen till detta är att AI-systemen förutser studentresultat, övervakar, ger feedback och pushar studenten till handling. I det bästa av scenarion kan däremot läraren ses som en partner till AI-systemen. Konferensartiklarna skriver inte mycket om lärare, för nästan allt fokus ligger på studenten. Det innebär att det är upp till systemen att ta plats på lärarnas bekostnad. Denna utveckling belyser vikten av att lärare och studenter får ta del av konsekvenserna av införandet av AI-system i skolan. / Dynamic AI-system are moving fast into the world of schools. This will affect education, teaching and learning and not at least the roles of teachers and students. That is why the aim of this study is to examine how teachers and students roles, according to the trends that emerge in the AIED 2021 Conference and the LAK21 Conference, are affected by AI-system in school. The starting point for this thesis are the studies that are presented during the two major conferences that are being held by the International Society of Artificial Intelligence in Education (IAED) and the Society for Learning Analytics (SoLAR) in 2021. In their research they both are focusing the use of AI in education. The societies are asking to be recognized as leading in their fields. Out of 187 articles published during the conferences 130 have fulfilled the criteria to qualify in this literature review. The articles have been categorified due to the AI-system that has been used and their influence on the role of the teacher and the student. The result demonstrates the roles of a more passive teacher and a more active student. The explanation for this is that the AI-system supervise, give feedback and push students into action. In the best of scenarios, the teacher can be seen as a equal partner with the systems. However, not much is being written about the teachers, it´s all about the students. Since the teacher is out of focus, it´s up to the system to fill that vacuum. This development makes it important that the consequences of the introduction of AI-systems in education are known to the concerned parties – the teachers and the students. / <p>Godkännande datum: 2022-06-03</p>