419
PROPOSED AI-BASED PERSONALIZED LEARNING MODEL ARCHITECTURE
FOR COLLABORATIVE DIGITAL LEARNING ENVIRONMENTS
Dao Thi Xuan Huong.
Faculty of Information Technology – Electrical Engineering,
Saigon South Polytechnic College.
Email: xuanhuong@nsg.edu.vn; Mobile: 84-0909546736
Abstract
One of the goals of education and learning in digital technology today; is to create for
learners a flexible and effective online learning environment, with AI technology that can
adapt to the needs and characteristics of each learner, while promoting active digital
interaction and collaboration among members in a learning group. This research topic
focuses on building a comprehensive architecture and identifying core components for a
personalized learning model, supported by artificial intelligence, in the context of
collaborative digital learning.
The proposed architecture will include functional modules such as collecting and analyzing
student data, with an intelligent AI system to provide personalized recommendations and
adjustments for learners, and tools to support groups in digital collaboration. The smooth
integration of these components aims to optimize the learning experience, improve
teamwork performance, and achieve better learning outcomes.
Keywords: Learning model, AI-Based Personalization, Collaborative Digital Learning.
Introduction
ackground
In the environment of the digital era with the explosion of technology and the drastic
transformation in the field of education, digital learning has become a mainstream trend,
bringing flexibility and access to vast open source knowledge. However, the majority of
current digital learning platforms still have limitations in meeting the diverse and unique
learning needs of individual learners, so they often provide mass learning content and paths.
At present, with the rapid development of AI technology, there are open opportunities to
personalize the learning experience, from analyzing learner data, recommending relevant
content to providing intelligent support. Besides, collaborative learning has proven to be an
effective method to enhance interaction, share knowledge, and develop essential soft skills
in the modern work environment. Therefore, the study of proposing an AI-based
personalized learning model architecture for collaborative digital learning environments
becomes extremely urgent. This research not only addresses the urgent need for a flexible,
adaptive, and highly interactive learning environment, but also makes full use of the
potential of AI to optimize learning efficiency for learners, bring efficiency to teaching, and
contribute to the innovation and quality improvement of the entire education system in the
context of strong digital transformation.
Topic’s Objectives:
420
The research paper gives the reasons and also sets the main goals of aiming at proposals to
solve the problems:
The strong development of digital learning with the need for personalization:
For each learner, there is interaction, group activities have been applied in the current digital
training process. With the mainstream trend in modern digital education, digital learning
with group interaction requires learners to have autonomy with individual learning plans,
flexibly combining their own and group activities. This helps learners expand their access
to: open knowledge, digital interaction, work-time management,... However, current digital
learning platforms often provide common learning content and pathways for all learners, not
meeting the unique needs and characteristics of each individual suitable for the general
activities of the study group. Therefore, the need for an adaptable and personalized learning
environment is becoming increasingly urgent to optimize the learning experience and
effectiveness. (Dillenbourg, 1999)
The development and potential of AI technology in education:
AI technology has been proving its superior ability to analyze big data, identify patterns,
make predictions, and make intelligent recommendations. Therefore, the application of AI
in learning can help better understand learners, automate repetitive tasks in planning,
provide personalized feedback, and create optimal learning paths. So the importance of AI
technology will help discover and exploit the potential of learners, to solve the problem of
personalization in the digital learning environment. Evidence of the potential of AI
technology in education has been researched from the world's leading universities such as
Harvard, MIT and Wharton indicating that learners who use AI often complete up to 12.2%
more work than the average, and up to 25.1% faster. at the same time, with a 40% higher
quality of work. (Fabrizio Dell'Acqua, Edward McFowland III, Ethan R. Mollick, Hila
Lifshitz-Assaf, Katherine Kellogg, Saran Rajendran, Lisa Krayer, François Candelon,Karim
R. Lakhani., 2023)
The need for collaboration in a digital learning environment:
The important thing now in learning in the digital environment is to bring many benefits
such as increased interaction, knowledge sharing, development of teamwork skills and
critical thinking. Therefore, the digital interactive learning environment needs to facilitate
effective collaboration between learners. This research aims to integrate collaborative
elements into an individualized learning model, creating a learning environment that is both
individualistic and socially interactive. (Dillenbourg, 1999)
The limitations of current personalized learning models:
Currently, many current personalized learning models focus mainly on content adjustment
and learning pace, with little emphasis on factors such as learning style, interests, personal
goals, and social interaction. (Nguyễn Thị Phương Lê, Lam Thị Loan, Vũ Thị Thuý Hằng,
2024)
In addition, the lack of user interaction with learning materials and lecturers, the ability to
understand and grasp basic knowledge of Vietnamese students is still weak and lacking; the
main reason is that students pay too much attention and "rely" on searching for knowledge
through technology. (Anh, 2025)
The integration of AI into these models is still in its infancy and has not yet been fully
exploited. This study aims to overcome those limitations by proposing a more
421
comprehensive architecture that effectively combines AI and collaborative elements.
(Nguyễn Thị Phương Lê, Lam Thị Loan, Vũ Thị Thuý Hằng, 2024)
Requirements for innovation and improvement of education quality in the context of
digital transformation:
Education is undergoing a strong digital transformation, requiring new, innovative and more
effective learning methods and models. This research contributes to the effort to innovate
education by proposing an advanced solution that takes advantage of the latest technologies
to improve the quality of teaching and learning. ―This is seen as an opportunity for
educational institutions to change the content and form of teaching to meet the individual
needs of learners, helping them adapt to the learning and working environment of the
future.‖ (Nguyễn Thị Phương Lê, Lam Thị Loan, Vũ Thị Thuý Hằng, 2024)
Importance of the research topic:
For Learners:
Optimize the learning experience: Through the personalization model, it will help
learners find the right learning content and methods for their needs, thereby increasing
their interest and motivation in learning.
Improve learning efficiency over time: through a learning pathway that is tailored to
the capacity and speed of each individual, helping learners master knowledge and
skills in the most effective way.
Comprehensive development: With the integration of digital collaboration elements,
it helps learners develop important soft skills such as communication, teamwork and
critical thinking.
Lifelong Learning Support: through a flexible development model and high
adaptability, it can support learners throughout their own learning and career
development.
For Instructors:
Providing effective teaching support tools: with AI tools that can help lecturers better
understand the needs and learning progress of each student, thereby providing timely
interventions and support.
Reduce the workload of management and evaluation: Because instructors can
automate a number of tasks such as tracking progress, grading multiple-choice
assignments, and providing initial feedback to learners.
Improving the quality of teaching: over a period of time, with an analytical dataset
from AI, it can provide useful information for lecturers to improve teaching methods
and design more suitable content for learners.
Create an interactive and engaging learning environment: through the integration of
intelligent collaboration and support tools from AI can create a dynamic and engaging
learning environment for learners.
For the education system:
Improving the quality and effectiveness of training: with a personalized learning model
has the potential to improve the output quality of the education system, better meeting
the needs of the labor market.
422
Promoting innovation in education: This research contributes to the discovery and
application of advanced technologies in the field of education, promoting innovation
and development of the industry.
Enhancing international competitiveness: An advanced and effective education system
will help the country improve its competitiveness in the international arena.
Contributing to the national digital transformation goal: This research contributes to
the application of digital technology, especially AI, in the field of education, one of the
important pillars of the national digital transformation process.
In conclusion, the research topic on AI-based personalized learning model architecture for
collaborative digital learning environment has scientific and practical significance, solving
current challenges and opening up new potentials for education in the digital era. Building
an effective model will bring practical benefits to learners, educators, and the entire
education system.
Theoretical basis and research points
Theoretical basis
Personalized Learning (PL)
Personalized learning is a learner-centered pedagogy in which the learning process is
tailored to each learner's needs, interests, learning style, learning pace, and personal goals.
(Barbara A. Bray, Kathleen A. McClaskey, 2016). Instead of a uniform educational
paradigm, PL creates separate learning pathways, empowers learners, and encourages
autonomy in the process of acquiring knowledge. (Heather Staker, Michael B. Horn, 2012).
The core elements of PL include:
Learner Profile: Collecting diverse information about learners, including
background knowledge, learning styles, interests, learning goals, and current
progress.
Flexible Learning Paths: Providing a wide selection of content, activities, and
assessment methods so that learners can choose the learning path that best suits them.
Pace-Based Learning: this allows learners to progress at their own pace, ensuring
insight before moving on to new concepts.
Targeted Feedback: Provide timely and specific feedback to help learners identify
strengths and weaknesses and adjust the learning process.
Learner Agency: Encourage learners to participate in the design and management of
their own learning process.
Artificial Intelligence in Education
Artificial Intelligence is increasingly being widely applied in education, bringing smart
solutions to improve teaching and learning efficiency. In the context of personalized
learning, AI plays a key role in:
Analyze learner data: AI is capable of processing large amounts of data about
learners such as: interaction behavior, learning outcomes, learning styles, etc.; to be
able to identify patterns, trends and provide insights into the needs and characteristics
of each individual. (Ryan S. Baker, George Siemens, 2012)
423
Recommend and personalize learning content: Based on data analysis, AI
algorithms can recommend learning materials, assignments, activities, and resources
tailored to each person's level, interests, and learning style.
Provide intelligent feedback: AI can automatically grade assignments, provide
detailed feedback, and suggest improvement steps for learners. Advanced AI systems
are also capable of analyzing natural language to understand and respond to complex
learners' questions.
Adaptive Learning Support: AI can adjust the difficulty of the content, the speed of
presentation, and the approach based on the learner's performance and understanding
level in real-time.
Supporting interaction and collaboration: AI can create tailored learning groups
based on common interests, levels, or goals, and support collaborative activities
through interactive analytics tools and provide suggestions.
Collaborative Digital Learning Environment (CDLE)
A collaborative digital learning environment is an online space where learners can interact
with each other, share ideas, solve problems together, and build knowledge through group
activities and joint projects. (Dillenbourg, 1999). Collaborative learning offers many
benefits, including developing communication skills, critical thinking, problem-solving, and
teamwork. In CDLE, digital tools and platforms support diverse forms of interaction such as
discussion forums, chat rooms, document sharing tools, and online group projects.
Theoretical framework and usage analysis
Proposed theoretical framework:
The architectural details of the AI-based personalized learning model for the digital
collaborative learning environment are proposed according to the tiered architecture
consisting of the following main layers or modules:
Data Layers: Collecting and managing learner data includes: personal information,
learning history, learners' own learning styles, interactive and cooperative activities
on digital platforms.
AI Layer: including AI modules responsible for analyzing data, making personalized
recommendations (in terms of content, activities, with a feasible roadmap); support
the formation of appropriate digital learning groups, and coordinate group
interactions.
Interface and interaction layer: in order to provide a friendly and suitable interface
for learners and study groups to interact with the system, improve the efficiency of
participation in individual learning activities and digital cooperation.
Service Floor: is the process of interaction between the system and learners, through
the process of providing services and processing to support digital learning for
learners. For example, a feedback system to support the line, a result evaluation tool,
a periodic communication tool to remind or encourage learners.
