The adaptive learning platform is based on the existing online learning platform MOOC and incorporates generative AI to solve two problems.

The first is to use generative AI to enrich course resources and assist in updating course content. MOOC has covered a lot of high-quality courses since its establishment, but there are some areas that have not yet been covered. We can use generative AI to combine the existing materials in related fields to quickly build related courses and complete the course categories. In addition, for the existing courses, there are also few expansion resources, insufficient related reading materials and insufficient practice questions, so we can use generative AI to enrich this part of the content.

The second is to use generative AI to enhance the interactivity between students and courses. Through automated analysis of the learner's knowledge level, interest and emotional state, AI dynamically generates customized learning content (e.g., practice questions, video explanations) and supports voice, image and other interaction methods. The design emphasizes “learner-centeredness” to generate differentiated lesson plans.

The platform is expected to promote educational equity and universalize quality resources through AI, especially in resource-poor areas. It is also expected to shift the role of teachers to that of facilitators and promote the development of a lifelong learning society.

Combined with my own learning experience from secondary school to university, as well as what I have seen and heard during my study visits and exchanges, I have found that there are great differences in educational resources in different countries and regions. The differences not only come from the material learning environment and laboratory equipment, but also from the teachers' qualifications and vision. The online education platform can provide us with an equal and easily accessible learning opportunity. Students can see information about courses in all disciplines through online education platforms, and they can also get guidance from high-quality teachers who are scarce in less economically developed areas. One of the values of education is to break down prejudice and promote equality, so I believe that the future model of education must be deeply integrated with online education platforms.

My hypothesis is to build an online education platform with the following functions:
1 With the support of generative AI, users input textbooks or teaching materials of a certain course, and the platform automatically searches through the Internet to generate a course containing teaching videos, extended reading, practice questions, and quizzes.
2 Students are free to discuss with the chatbot any questions they may have during the course.
3 At the end of the course, assessment questions are randomly generated for students, and a certificate of completion is issued to those who pass.

I think the existing AI technology can already meet the requirements and can be built and tested on a MOOC, which I believe can be done well after a few rounds of optimization.

Technical Implementation:
The platform will build an adaptive learning engine based on the Transformer model, which dynamically generates personalized learning paths by analyzing students' learning behaviors (e.g., speed of answering questions, types of errors). For example, using Stable Diffusion technology, textbook content can be transformed into a 3-minute situational skit (e.g., demonstrating physical mechanics with Journey to the West characters). The real-time feedback system, on the other hand, completes grammar correction of essays in 0.3 seconds with a lightweight deployment of the AI correction engine, and labels specific suggestions for improvement.

Key resources：
The project needs to reduce the cost of computing through AWS + local servers. For educational data, it needs to acquire a library of labeled exercises and hire a team of teachers to review the AI-generated content. The development team needs to include AI engineers, curriculum designers and UX researchers.

Market Demand Verification:
Analyse the feedback from students using MOOC using questionnaires.

Core Functions：
（1） Intelligent diagnosis: accurately locate knowledge weaknesses,
（2）Dynamic path: real-time adjustment of learning routes like navigation software, automatic switching of knowledge types when fatigue is detected,
（3）Scenario-based generation: customize cases based on students' living environments.

User Experience and Effectiveness Evaluation：
Set up a double content review mechanism, AI-generated courses are subject to teacher spot checks (5% sample size per day), and model retraining is automatically triggered when the error rate exceeds 2%. Key indicators include:
（1）Learning efficiency: Knowledge mastery time is reduced to 60% of traditional teaching,
（2）System response: content generation delay of no more than 1.2 seconds,
（3）Continuous use: more than 80%of users actively use it more than 4 times per week.

Core Functional Modules
(1) Intelligent Weakness Diagnosis
Students log in and do 10 probing questions like assembling circuit diagrams by drag-and-drop to locate weak points within 5 minutes
(2) Dynamic Lesson Generation
Based on the diagnostic results, two types of content are generated in a hybrid way: video micro-lessons/interactive experimental dialect exercise packages.
(3) Real-time feedback system: essay correction/attention monitoring.

1 Generative AI rebuild the course resources
The platform leverages generative AI to overcome the limitations of traditional MOOC resources, rapidly creating courses in emerging fields and supplementary materials like adaptive exercises or multilingual handouts. By continuously tracking academic developments, it enables automatic content updates, addressing gaps in coverage and timeliness, and building an efficient human-AI collaborative content ecosystem.

2 AI improve the effect of learning interaction
The platform employs multimodal interactions (voice, handwriting) to dynamically adapt to learner needs, generating personalized learning paths and assisting teachers in designing precise instructional strategies. This drives online education to evolve from "mass-scale dissemination" to "mass-scale + precision" advancement.

At certain times, there may be a surge in the number of users, such as a sudden influx of a large number of students during the school season, which requires the system to be able to automatically increase the cloud server. For the problem of unstable network in economically underdeveloped areas, the platform needs to convert the course content into a streamlined version in advance and cache it on the local server to ensure smooth learning even when the network is jammed. In addition, the future also needs to develop mobile applications, no network can also watch the cached content, and when the network is restored to automatically synchronize the learning records, so as not to let the progress of the “offline”.

For sustainability, we need to consider the power consumption of AI computing, and combine solar/wind energy resources to promote green AI. on the other hand, we can turn learning into a game - for children who like soccer, the math problem will be turned into a calculation of the success rate of the shot; students can also become “small editors of the curriculum”, and discover the AI-generated content. Students can also become “Little Curriculum Editors” and find errors in the history lessons generated by AI, and mark the changes to save points for scholarships to enhance students' interest in learning.

The generative AI-supported adaptive education platform solves the problems of uneven distribution of educational resources and homogenization of education models through generative AI technology. For economically underdeveloped regions, the platform provides customized resources to lower the threshold of access to quality education; at the same time, it rapidly generates courses in emerging fields to fill the gaps in disciplines neglected by traditional MOOC. By dynamically generating personalized learning paths, it promotes educational equity and inclusion.

Core indicators include quantitative data such as the percentage of users from groups in economically underdeveloped regions, course completion rates, access rates in economically underdeveloped regions, and qualitative assessments such as user feedback. Through a real-time feedback system, the platform transforms social demand into measurable incremental value and continues to promote educational equity and inclusion.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.