extended essay academic writing topic transform education with ai the rise of intelligent tutor systems

Adaptive Learning Paths Based on Individual Student Progress.... These systems tailor the educational experience to individual student needs and preferences, functioning similarly to a h

MINISTRY EDUCATION & TRAINING NATIONAL ECONOMICS UNIVERSITY

EXTENDED ESSAY ACADEMIC WRITING

TOPIC: “TRANSFORM EDUCATION WITH AI: THE RISE

OF INTELLIGENT TUTOR SYSTEMS”

Student performing: Hoang Thi Mai Anh Student ID number : 11230731

Class: Business Analytics 65 Supervisor:

Ha Noi, October 2023

DECLARATION OF AUTHENTICITY

I hereby affirm that this work is the result of my personal research and effort All sources

of information, data, and referenced materials used in this thesis have been clearly and

accurately cited | assert that this thesis does not contain any part copied from the work

of others without proper acknowledgment or citation

Hoang Thi Mai Anh

ACKNOWLEDGMENTS Firstly, I would like to express my gratitude to Professor Nguyén Hong Sơn for your dedicated teaching and guidance throughout the course Thanks to your detailed lectures and attentive instruction, I have overcome the challenges encountered in completing my essay

Next, I extend my sincere thanks to the faculty of the National Economics University Their knowledge transfer has helped me build a strong foundation for my studies

Additionally, 1 must acknowledge my family, friends, and relatives who have been my steadfast support, providing emotional strength, encouragement, and assistance

throughout this journey

Lastly, I recognize that with my modest level of knowledge and capabilities, this essay may not be free from shortcomings I humbly request Professor Nguyén Hong Son to overlook these and provide constructive feedback for my continuous improvement

Hanoi, October 20, 2023

Hoang Thi Mai Anh

Table of Contents

DECLUARATION OF AUTHENTTICHTTY - - LH HH KH KH TT TH KH

_ 0.0920) 6)i10 01100 66G (d4

II 00.(0))0)009525555 4 ‹‹-44 3

II BENEFITS OF AI TUTOR SYSTEM IN EDUCA TION - Q HH

1 Personalized Learning Êxperienee - - - acc ni nh cv 1.1 Customization of Learning Materials and Pace eee eee 1.2 Adaptive Learning Paths Based on Individual Student Progress - -

2 Immediate Feedback and Adaptability of AI ITS in Education 00 ee EM) r0 e 3.1 Analysis of Student Data for Idendfying Learning Patterns - 3.2 Utilizing Data to Enhance Teaching Methods and Curriculum Design

Il CHALLENGES IN IMPLEMENTING AI TUTOR SYSTEMS ee

IV Future Prospects and Innovations in Al-Driven Education 2::ce:cceceeceeeeeeeeeeeeneeeeeeees

V CONCLUSION 30040000) 2117 .aảaaÃä4£43

I INTRODUCTION Education, a crucial sector of society, faces numerous challenges, such as accessibility, the implications of events like the COVID-19 pandemic, and financial barriers These issues have catalyzed the exploration of technological solutions, notably the integration

of Artificial Intelligence (AI) Al's significant role in education is increasingly

recognized, offering innovative approaches to address these challenges (Ahmad et al.,

2021)

Intelligent Tutoring Systems (ITS), as a manifestation of Al in education, represent a sophisticated method of instruction These systems tailor the educational experience to individual student needs and preferences, functioning similarly to a human tutor but in a digital context The application of Al in ITS ranges from personalized content delivery and assessment of student knowledge to interactive problem-solving The integration of various AI tools and platforms within ITS highlights the diverse potential of AI to

enhance both learning and teaching

The rapid adoption of Al in education underscores its pivotal role in transforming the field Predictions suggest a substantial increase in Al technology use in educational

settings, indicating a recognition of Al's importance in education However, despite the growing trend, many educators remain unaware of Al's full capabilities The increasing presence of AI, coupled with the need for greater awareness among education

professionals, marks a significant evolution in the intersection of Al and education

(Ahmad et al., 2021)

II BENEFITS OF AI TUTOR SYSTEM IN EDUCATION

1, Personalized Learning Experience

1.1, Customization of Learning Materials and Pace

AJ in education allows for the customization and personalization of curriculum and

content, aligning them with students! needs, abilities, and capabilities This customization fosters better uptake and retention of information, thereby enhancing the overall quality

of learning (Chen, L., Chen, P., & Lin, Z., 2020)

Intelligent Tutoring Systems (ITS) offer personalized instruction to help learners master knowledge and skills These systems are superior to typical computer-based training as they adapt to the individual learner at a fine-grained level (Morgan, Hogan, Hampton, Lippert, & Graesser, 2020)

1.2 Adaptive Learning Paths Based on Individual Student Progress

The learner model within AI learning systems is crucial for enhancing independent

learning capabilities It is based on learners’ behavior data and includes analyzing their thinking and capability to assess their learning abilities Knowledge analysis is mapped

to gauge learners’ mastery of knowledge This model forms the basis for instructors to tailor teaching strategies and actions, enhancing education by always being prepared to offer aid from the tutoring model’s built-in teaching theories Machine learning, learning analytics, and data mining are closely related technologies pivotal in education Learning analytics focuses on data from student characteristics and knowledge objects from the learner model, tailoring educational methods to individual learners' needs and abilities This includes intervening with students at risk or providing feedback and instructional content Al-based educational data mining aims to develop association rules and offer knowledge objects to meet students! personal needs, thereby improving the learning process and knowledge mastery AI in education has led to the development of various platforms and applications that foster the customization and personalization of content Applications like Knewton make real-time recommendations for students based on

learning styles deciphered by machine learning algorithms, customizing course materials

to learners’ needs Similar capabilities are seen in platforms like Cerego,

Immersivereader, and CALL These platforms improve the learning experience across different education levels (Chen, L., et al., 2020)

ITSs employ intelligent algorithms that adapt to the individual student's needs,

encompassing the mastery of subject matter, cognitive skills, motivation, and sometimes emotions This adaptivity is sensitive to the student’s prior performance, student model

of the subject matter, and other psychological characteristics, making ITSs more

effective than traditional computer-based training (Morgan et al., 2020)

The standard architecture of ITS includes the domain model, the student model, the tutor model, and the user interface The domain and student models work together to provide information to the tutor model, which determines the subsequent actions of the tutor, thus enabling adaptive learning paths (Sottilare, Graesser, Hu, & Holden, 2013; Woolf, 2009)

2 Immediate Feedback and Adaptability of AT ITS in Education

Intelligent adaptive learning systems, integral to contemporary educational technologies, are specifically engineered for adaptability and interactivity These systems astutely evaluate and react to a range of learner activities, including providing answers and

decision-making processes The feedback offered by these systems varies extensively, from simple confirmations of correctness or incorrectness of responses to comprehensive explanations aimed at enhancing learner performance Additionally, these technologies empower learners by presenting them with choices in their learning paths, thereby

promoting self-regulation in their educational journey

The architecture of Intelligent Tutoring Systems (ITS) features a learner model that

reflects various aspects of the learner, such as knowledge, affect, motivation, and other psychological characteristics This model is frequently juxtaposed with the domain

model to track student progress ITSs function is based on two primary loops: the outer loop, which identifies the optimal task for the learner, and the inner loop, which manages actions within the task, including feedback and hints This dual-loop mechanism allows

ITSs to adapt their instructional methodologies to the individual learner's requirements, often integrating emotional cues from verbal and non-verbal interactions

A significant innovation in ITSs is the implementation of stealth assessments, which unobtrusively gather data while the learner engages with the content, thus assessing progress seamlessly and without disturbing the learning flow This method allows for a comprehensive assessment that includes cognitive, metacognitive, and social measures, further enhancing the ITS's ability to tailor its approach to the learner's specific needs (Morgan, B., Hogan, A., Hampton, A J., Lippert, A., & Graesser, A C., 2020)

3 Data-Driven Insights

3.1 Analysis of Student Data for Identifying Learning Patterns

Al-powered algorithms play a pivotal role in examining student data to craft

personalized learning experiences This analytical process enables more effective and efficient learning, significantly reducing the time and resources typically expended in conventional teaching methodologies (Lin, Huang, & Lu, 2023)

3.2 Utilizing Data to Enhance Teaching Methods and Curriculum Design

In the realm of adaptive learning within tutoring systems, AI is instrumental in

personalizing the learning experience Adaptive learning, an approach focused on

maximizing each student's learning potential, leverages Al to analyze learning patterns

By scrutinizing these patterns, Al aids in formulating a customized learning process that aligns with each student's capabilities, thereby optimizing academic performance This intelligent application of AI enables educators to discern students’ strengths and

weaknesses, subsequently adapting their teaching strategies to cater to individual student needs (Lin et al., 2023)

III CHALLENGES IN IMPLEMENTING AI TUTOR SYSTEMS

Implementing Al Tutor Systems presents unique challenges to researchers and

practitioners in both the computer science and education fields Developing intelligent tutoring systems and adaptive learning systems requires not only advanced programming skills but also an understanding of how to simulate human intelligence in educational settings This includes integrating the knowledge and experience of human tutors into the systems to make informed judgments and decisions, tailored to address individual

learners' problems and enhance their learning experience The complexity of AI in

Education (AIED) stems from its dependency on technology and its interdisciplinary nature, demanding a comprehensive understanding of both Al's role in education and its operational mechanisms Without this understanding, the effective implementation of AIED applications and the exploration of new research avenues within this field become significantly more challenging (Hwang, G.-J., et al., 2020)

One significant challenge in the implementation of Intelligent Tutoring Systems (ITS) is their limited impact on education and training worldwide Several factors contribute to this issue Firstly, the development of ITS is costly, and until recently, the necessary computing power was also expensive Additionally, the genesis of the field primarily among artificial intelligence researchers, rather than education specialists, has led to a focus on the advancement of AI algorithms rather than the educational effectiveness of these systems This has resulted in ITS being evaluated more on their Al capabilities - specifically, their ability to interpret and respond to student behaviors - rather than their cost/benefit ratio in educational terms (Corbett, A T., Koedinger, K R., & Anderson, J

R., 1997)

Furthermore, for ITS to make a significant impact on the educational and training

systems, there needs to be a shift in the evaluative focus from AI sufficiency to

educational impact Although there is a growing trend towards this shift, it is still in its nascent stages For instance, at recent International Conferences on Intelligent Tutoring Systems, only a quarter of non-invited papers included any form of empirical evaluation Among these, a mere 2.5% assessed the effectiveness of ITS by comparing student

performance in computer-based learning environments to other learning environments

This highlights the need for a more profound emphasis on the educational impact at every stage of ITS development, deployment, and assessment (Corbett et al., 1997)

IV Future Prospects and Innovations in Al-Driven Education

1 Al-Driven Personalized Curricula

The evolution of Intelligent Tutoring Systems (ITS) points towards a future where Al- driven personalized curricula become the norm ITS can tailor entire curricula based on individual student progress and interests, thereby enhancing the learning experience By integrating AI systems with curriculum planning, a seamless and more effective

educational experience is achieved This individualized approach is particularly vital in areas like problem-solving in various domains, as well as in supporting activities like studying examples and exploring interactive simulations (Conati, 2009)

2 Emotional Intelligence in AI Tutors

Emotional intelligence in Al tutors is a burgeoning field with significant potential to transform educational experiences The development of AI systems capable of

recognizing and responding to students’ emotions can lead to enhanced engagement and motivation Such emotionally intelligent Al tutors can adapt to the emotional states of students, offering personalized support and feedback that resonates on a more personal level (Conati & Maclaren, 2009)

3 Collaborative Learning Environments

Al technology also holds promise in fostering collaborative learning environments AI- supported group activities and collaborative projects can enhance the development of teamwork and communication skills By facilitating Al-driven collaboration, students can benefit from the synergy of group learning while still receiving individualized

guidance and support This approach aligns with the broader goals of ITS to provide individualized support for a variety of educational activities (Conati, 2009)