Higher Education Category Submission

506 – WRITE! – Writing Reinforcement and Intelligent Tutoring Engine

Our project aims to revolutionize English writing education for primary and secondary school students by addressing the limitations of traditional teaching methods. Existing approaches often rely on standardized materials, such as prearranged guiding questions and word banks, which fail to accommodate the diverse learning needs of students. This disconnect between instructional materials and individual learners hinders effective skill development and engagement.

To tackle this challenge, our team has developed an innovative, AI-powered writing platform that provides personalized and dynamic guidance tailored to each student's abilities and progress. Leveraging advanced Large Language Models (LLMs), the platform offers customized assistance, such as vocabulary recommendations, grammar corrections, and contextualized feedback, creating an adaptive learning experience.

The design of our solution emphasizes student-centric learning, integrating data-driven insights into a responsive and intuitive interface. The platform supports students through structured writing stages—planning, drafting, revising, and sharing—while fostering collaboration through peer interaction and co-writing features. This approach not only enhances individual writing skills but also builds a sense of community and motivation among learners.

The potential impact of our project is profound. By addressing key pain points like vocabulary gaps, grammar confusion, and disengagement, our platform empowers students to overcome writing challenges and achieve measurable improvement in their English proficiency. Teachers also benefit from real-time visibility into students' progress, enabling timely interventions and targeted feedback.

Ultimately, our solution redefines the writing classroom by transforming writing into an engaging, supportive, and collaborative experience. By combining cutting-edge AI technology with innovative educational design, we aim to bridge the gap between diverse learner needs and effective teaching methods, fostering equitable access to quality education and equipping students with the skills necessary for academic and professional success.

Our project originated from observing the significant challenges faced by students and teachers in English writing education, particularly in ESL (English as a Second Language) contexts. Students often encounter barriers such as limited vocabulary, grammatical difficulties, and redundancy of ideas, which prevent them from expressing themselves effectively. Research shows that inadequate vocabulary leads to oversimplification or misuse of words, while grammatical issues, such as tense agreement and sentence structure, result in incoherent writing (Farooq et al., 2012). Furthermore, redundancy of ideas weakens the clarity and impact of students’ messages (Peter & Govindarajan, 2021).

Teachers, on the other hand, struggle to address these issues due to the diverse proficiency levels within classrooms. Some students grapple with basic sentence formation, while others are ready for advanced writing tasks, making differentiated instruction necessary but time-consuming. This challenge is further compounded by limited resources and an increased workload, which reduces the capacity for personalized feedback and meaningful engagement (Moses & Mohamad, 2019).
Our thought process focused on bridging this gap by leveraging Generative AI to create a personalized, adaptive, and scalable solution. The platform aims to provide real-time writing support tailored to individual student needs, including vocabulary recommendations, grammar corrections, and organization of ideas. By incorporating features such as collaborative writing and structured feedback, the platform fosters engagement and encourages progressive learning.

The hypothesis underlying our project is that a data-driven, AI-powered platform can effectively enhance ESL writing education by addressing the specific pain points of students and teachers. This solution is designed to improve learning outcomes through personalization and accessibility while reducing teachers' workload. By addressing key challenges in writing education, the platform has the potential to make a lasting impact on how students learn and teachers teach, ensuring broader scalability and sustainability in diverse educational settings.

Our solution utilizes AI-driven tools to assist students in writing with real-time, personalized advice. The core functionalities of the app are:

1. Guiding Questions for Brainstorming: The app generates five types of guiding questions (General, Literal, Reflective, Interpretive, Critical) to help students expand their ideas and think critically about their topics.
2. Vocabulary Bank: Using AI, the app suggests categorized vocabulary based on the writing topic, enabling users to enhance their expression and professionalism. Students can save words to a vocabulary list for future use and access definitions through a built-in dictionary.
3. Advice on Word Choice: The app analyzes the entire passage, highlights areas for improvement, and provides alternative phrases to refine clarity, style, and impact.
4. Grammar Check: The app scans for grammatical errors, including sentence structure, punctuation, spelling, and subject-verb agreement, offering correction suggestions to improve overall readability.
5. Save and Revise: Students can save their drafts and vocabulary lists for future reference and revisions, encouraging iterative learning.

To implement the solution, we leverage Generative AI models and Natural Language Processing (NLP) to power the writing assistant. The app will be hosted on a cloud infrastructure to ensure scalability and accessibility. Resources include AI development tools (e.g., OpenAI API), labeled datasets for fine-tuning, and a user-friendly interface designed for seamless interaction.

Validation of market demand will involve pilot testing in schools and gathering feedback through surveys with students and educators.

Positive user experience is ensured through an intuitive interface, categorized suggestions, and actionable recommendations. Performance metrics, such as improvement in writing quality (pre/post-assessments), user engagement (time spent on tasks), and error reduction rates, will measure the app's effectiveness in enhancing writing skills.

Functional Architecture and Workflow

Our project will bebuilt on a modular, cloud-based architecture to ensure scalability and adaptability. It consists of:

1. User Interface Layer: A responsive, intuitive interface (web and mobile) providing tools like text editing, brainstorming prompts, vocabulary suggestions, grammar checks, and word choice advice.
2. Application Logic Layer: AI-driven modules for brainstorming, vocabulary generation, text analysis, and grammar correction, with a personalization engine to adapt to user progress.
3. Data Layer: A secure, cloud-based database for storing user drafts, saved vocabulary, and progress history.
AI and NLP Layer: NLP and ML models for grammar checking, vocabulary generation, and text analysis, fine-tuned to meet the needs of ESL learners and students with special education needs (SEN).

Workflow: Users input text or topics → AI processes the input → Features (e.g., brainstorming tool, grammar check) analyze and offer suggestions in real-time → Data is saved for tracking progress.

Implementation of Innovative Features

Brainstorming Tool: GPT-based models generate categorized guiding questions to help organize ideas.
Vocabulary Bank: Pre-trained word embedding models suggest topic-relevant words and allow saving for future use.
Grammar Check and Word Choice Advice: NLP algorithms analyze drafts for errors and offer corrections with explanations.
Accessibility: Multilingual support, simplified UX, and progress tracking for SEN and ESL students.

Design and Development Timeline

Planning: 1 month – Requirements gathering, prototyping.
Development: 3 months – Build core features, implement AI tools.
Testing: 1 month – Pilot testing with diverse student groups.
Deployment: 1 month – Launch with feedback monitoring.

Performance Metrics

Writing Quality: Pre- and post-assessments of student writing.
Engagement: Usage data (time spent, features used, revisions).
Inclusivity: Adoption rates among ESL and SEN students.
Satisfaction: User feedback and surveys.

Our proposed platform is a collaborative writing tool specifically designed for primary and secondary school students who struggle with writing in English. Unlike traditional classroom approaches that emphasize rote memorization or isolated writing tasks, our solution introduces a co-writing environment that integrates planning, peer interaction, personalized vocabulary support, and essay sharing into a single digital experience.

The innovation lies in addressing multiple pain points—lack of vocabulary, grammar confusion, mother tongue interference, and disconnection from the writing process—through an interactive, student-centered platform. Students are guided through structured writing stages: idea planning, real-time co-authoring, and post-writing feedback. The inclusion of a saved vocabulary bank essay section allow for progressive learning, while essay sharing builds community and motivation.

Creatively, the platform mimics social and collaborative tools that students are already comfortable with, making the experience feel intuitive and enjoyable. We’ve reimagined the writing classroom as a live, peer-powered ecosystem that enhances both student autonomy and teacher visibility. Teachers can track drafts, suggest revisions, and assign group writing tasks—solving the time and feedback limitations highlighted in academic research (e.g., Moses & Mohamad, 2019).

By combining educational scaffolding with modern UX design, our tool doesn’t just improve writing outcomes—it transforms writing into a shared, empowering experience. This redefinition of how writing is taught and practiced is what makes our solution both innovative and deeply aligned with the actual needs of students and educators.

To ensure scalability, our solution is designed to leverage a robust cloud-based infrastructure with scalable computing capabilities. By utilizing platforms such as Amazon Web Services (AWS) or Google Cloud, we can dynamically allocate resources to meet increasing user demand. The core AI functionalities, such as grammar checking, vocabulary suggestions, and word choice advice, will be powered by API-based microservices that can scale independently. To address potential bottlenecks, we will implement load balancing to distribute traffic efficiently and optimize database queries for faster response times. Caching frequently accessed data, such as saved vocabulary and user preferences, will further reduce latency and improve performance.

Environmental sustainability is a key consideration in our solution. By optimizing AI model performance and using energy-efficient cloud services, we aim to minimize the carbon footprint of our operations. Additionally, we will prioritize serverless architectures where possible, as these consume fewer resources by scaling automatically based on usage.

To foster long-term user engagement, the platform will include features that encourage continuous learning, such as progress tracking, personalized feedback, and gamified rewards for achieving writing milestones. Regular updates, based on user feedback and advancements in AI technology, will ensure the app remains relevant and effective.

Adapting to evolving user needs will involve maintaining an open feedback loop with users through in-app surveys and analytics. The modular design of the platform allows for seamless integration of new features, such as support for additional languages, accessibility tools, or advanced analytics for educators.

By combining scalable architecture, energy-efficient practices, and user-centered design, our solution will not only meet the demands of a growing user base but also ensure long-term impact, user satisfaction, and environmental responsibility. This comprehensive approach positions the platform for both immediate success and sustained relevance in the future.

Our solution addresses critical social issues in education by promoting equitable access to high-quality writing assistance. Many students, especially ESL learners, face challenges in developing writing skills due to limited access to individualized support and resources. By leveraging AI to provide personalized and dynamic writing advice, our solution helps bridge this gap, empowering students from diverse backgrounds to improve their writing proficiency.

For students with SEN, the platform offers features such as guiding questions for brainstorming, vocabulary banks, word choice advice, and grammar checks, which are designed to simplify complex tasks and support their unique learning needs. Accessibility features, such as an intuitive interface, flexible pacing through draft-saving options, and visual aids, create a supportive environment where these students can thrive.

For students from underprivileged backgrounds, the platform ensures affordability and accessibility by leveraging scalable cloud-based technologies and requiring minimal resources (e.g., a basic internet connection) to operate. This reduces barriers to entry, enabling students from low-income families to benefit from tools that are typically only available through private tutoring or advanced educational programs.

To measure social impact, we will track:
1. Writing Improvements: Progress in writing quality through pre- and post-assessments.
2. Engagement Metrics: Usage data, such as time spent on tasks, revisions made, and vocabulary improvements.
3. Equity and Inclusion: Adoption rates across socioeconomic, linguistic, and educational backgrounds, including students with SEN.
4. User Feedback: Insights from students, educators, and parents to continually refine the platform.

By addressing the needs of students with SEN and those from disadvantaged backgrounds, our solution fosters inclusivity, levels the playing field, and empowers underrepresented groups to improve their writing skills and thrive academically. This commitment to equity ensures long-term social impact and aligns with broader goals of accessibility and educational fairness.

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.