Higher Education Category Submission

494 – EduMindSpark

The name of our project is EduMindSpark, symbolizing brainstorming and inspiring teachers to create effective lessons. This reflects the core innovation of our initiative: to support frontline teachers in engaging in a collaborative, brainstorming-like process of deeper thinking during interdisciplinary lesson planning.
The design concept of EduMindSpark is rooted in educational theories related to teacher learning and the Six Thinking Hats framework. We aim to address the current challenges faced by frontline teachers, who often recognize the importance of education and research but feel a disconnect arising from the experience of "high investment with low return." Our hope is that this project will provide opportunities for brainstorming and deep thinking for teachers seeking innovative breakthroughs but lacking the necessary resources for collaboration.
The technical foundation of the project is based on Microsoft’s open-source experimental Python library, TinyTroupe. In this framework, "TinyPersons" are AI-powered agents, each with unique personalities, interests, and goals. These agents can interact with one another and their environment, responding to various stimuli. "TinyWorld" refers to the virtual space where TinyPersons exist and interact.
The TinyTroupe framework includes a real-time feedback mechanism that dynamically adjusts character behaviors and decisions based on simulation results. This feature supports multi-level character interactions, allowing for complex dialogue between different characters. In TinyWorld, teachers can customize various types of TinyPersons according to their specific teaching and research needs, facilitating multiple rounds of discussion and exchange among virtual educators.
This project proposes an interdisciplinary collaborative lesson planning approach that combines one real teacher with multiple virtual teachers. By doing so, it aims to provide innovative thinking support, enhancing interdisciplinary collaboration and deeper thinking in lesson planning.

With the rapid development of generative AI, there are numerous applications for AI in education. Currently, much of the generative AI research focuses on enabling large language models to create lesson plans for teachers. Teachers can either use these generated lesson plans with slight modifications or treat them as inspiration, which often leads to a superficial engagement with the material. Many teachers do not fully recognize the convenience that generative AI can offer in educational settings.
In late 2024, when TinyTroupe was launched, the business community conducted in-depth research on its applications for product design. TinyTroupe emphasizes 'brainstorming' and 'deep communication,' making it particularly suitable for addressing challenges in teaching and research. Teachers are seeking innovative breakthroughs but often lack the collaborative resources to achieve them. Thus, I believe this project holds significant practical value in educational contexts. The assumption for this project is that teachers can utilize EduMindSpark to brainstorm with one another, sparking innovative ideas in teaching and research. This collaborative effort can lead to the development of high-quality courses and improved teaching efficiency.
Given TinyTroupe's strong performance in commercial design and the pressing needs of frontline teachers for brainstorming and innovation, I am confident that this project aligns well with current realities.Additionally, as a team of graduate students from the School of Educational Technology at Beijing Normal University, we possess a high level of professionalism in both theoretical guidance and technical learning prior to the project's implementation. We are also well-equipped for actual code deployment and iterative design throughout the project. Therefore, I believe this initiative is likely to be successful.

We will harness the power of Microsoft's open-source Python library, TinyTroupe, to bring our solution to life. Utilizing this technology, teachers will be able to predefine various roles for TinyPersons, engaging in numerous rounds of discussions and exchanges to foster interdisciplinary collaborative lesson planning and brainstorming sessions.
The resources required for this project encompass two main components: first, a technical team tasked with the secondary development of the TinyTroupe library and ongoing project maintenance; second, a team of educational experts to provide theoretical and practical guidance for the project.
The core functionality of our platform is to support teachers in interdisciplinary collaborative lesson planning and facilitate profound thinking. This includes four key features: a user-friendly interface with straightforward operation; a wealth of inspiration for teachers in lesson preparation; integrated teaching resources and case studies; and real-time feedback and optimization suggestions.
The effectiveness of our performance will be gauged through two evaluation metrics: first, a satisfaction survey among teachers regarding the generated lesson plans; second, the rate of conversion of teaching outcomes, which measures the proportion of teachers implementing the project's outcomes in their actual teaching practices.

none

EduMindSpark represents an innovative and creative solution to the issues outlined in the theme in several key aspects：
1.Facilitating Deep-Level Thinking: EduMindSpark is designed to stimulate teachers’ creative inspiration through a brainstorming-like process. This approach is innovative as it encourages deep-level thinking during the interdisciplinary collaborative lesson planning process, which is a significant departure from traditional teaching research methods.
2.Applying Educational Theories: By integrating the Six Thinking Hats theory into our project design, we offer a structured framework for thinking that enhances the creativity and effectiveness of collaborative teaching research. This application of educational theory is a creative solution to the common challenge of “high investment-low sense of achievement” experienced by teachers.
3.Customizable Role-Playing: The use of the TinyTroupe library allows teachers to predefined different roles (TinyPersons) based on their specific teaching research needs. This role-playing feature is an innovative way to simulate diverse perspectives and encourage in-depth discussions, which can lead to more innovative and effective teaching strategies.
Our project demonstrates innovation and creativity in the following ways, which enhance its effectiveness in addressing user challenges:
1.Personalization and Adaptability: EduMindSpark allows teachers to customize roles and discussion scenarios according to their individual needs, which enhances its effectiveness in addressing user challenges through a personalized approach.
2.Real-time Feedback and Optimization Suggestions: The real-time feedback and optimization suggestions component of EduMindSpark is a creative approach to continuous improvement. It ensures that the platform evolves with the development of user needs, becoming more effective in solving their challenges over time.

To meet growing user demand, the project leverages open API integration to embed deeply into teachers' lesson preparation, enabling flexible expansion of cross-disciplinary role libraries. For example, geography teachers can quickly integrate history and politics role libraries via APIs, simulating interdisciplinary lesson planning. Third-party developers or institutions can create custom AI role tools for specific needs (e.g., STEM courses), enhancing collaboration. The system’s modular design decouples core lesson functions (e.g., lesson plan generation) from extension modules (e.g., knowledge bases), allowing dynamic resource allocation through a distributed architecture. This prevents performance bottlenecks during high-concurrency collaboration, ensuring smooth multi-role lesson preparation.
For sustainability, the project adopts a user-driven iteration mechanism aligned with teachers' needs. Teachers can submit feedback (e.g., balancing content depth with engagement) through the built-in system. By applying educational theories (e.g., de Bono’s "green hat" creative thinking), the AI role behavior logic is updated quarterly to improve collaborative suggestions. The project also establishes a teacher community platform, where teachers can upload interdisciplinary cases to a shared repository for reuse and improvement. A points-based incentive system encourages contributions: teachers sharing quality cases or collaborating on lesson design (e.g., co-developing a "data mapping" activity integrating IT and geography) can unlock advanced features, such as personalized AI role training.
By supporting cross-disciplinary collaboration with technical scalability, implementing a dynamic iteration mechanism, and fostering a resource-sharing community, the project empowers innovative lesson preparation while building a sustainable co-evolution system.

5.1 Social Impact and Responsibility of the Project
EduMindSpark addresses the core contradiction of "high input–low sense of gain" in traditional collaborative lesson planning through a human-AI co-teaching model, promoting equity and inclusivity in educational digital transformation. The project focuses on the following social values:
1.Empowering Educational Equity: By leveraging TinyTroupe technology, the project builds a virtual lesson preparation community that breaks down regional and disciplinary barriers to quality educational resources. This provides teachers in underdeveloped areas with cross-disciplinary expert support, lowering the threshold for participating in innovative lesson preparation.
2.Advancing Cognitive Democratization: Through AI role dialogues guided by the Six Thinking Hats framework, the project mitigates the dominance of authoritative voices in lesson preparation groups. It ensures new teachers and marginal subjects have equal space for expression, promoting procedural fairness in knowledge production.
3.Enhancing Teachers' Professional Development: By simulating dynamic student feedback and fostering cross-disciplinary thinking, the project helps teachers overcome experience-based limitations, systematically improving instructional design creativity and complex problem-solving abilities, ultimately contributing to better classroom outcomes.
5.2 Impact Measurement and Response Mechanism
The project will establish three tiers of quantitative indicators:
1.Process Indicators: Participation rate in cross-school lesson preparation communities, frequency of interdisciplinary collaboration, and depth of AI role interaction.
2.Outcome Indicators: Scale of the innovative teaching case repository, assessment of teachers' instructional design improvement, and evaluation of students' higher-order thinking development.
3.Social Indicators: The spillover effect of lesson preparation outcomes in resource-limited schools and the regional educational equity index.
Through an embedded lesson preparation behavior analysis module and quarterly needs assessments, the project will dynamically optimize AI role configurations and develop localized toolkits for different regions, addressing the diverse needs of urban and rural teachers. By combining open-source technology with teacher practice communities, the project aims to build a sustainable and evolving educational innovation ecosystem.

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.