Open Category Submission

256 – Enhancing Student Competency in Writing User Requirements and UML Diagrams through Generative AI Tool

Drawing accurate UML diagrams and writing accurate user requirements are two important software engineering skills that students frequently find difficult to master. Traditional education methods are lack of engagement and lack of personalized feedback. This study try to solve this issues by applying generative artificial intelligence tools to build an interactive learning environment through which students can practice writing user/system requirements and drawing UML diagrams and receive real-time feedback along with personalized guidance.
It proposes a multi-layered to enhance students' skills in writing user requirements and creating UML diagrams. (1) User interface layer which present a user-friendly and responsive interface built with front-end frameworks. It allows students to engage with the platform, input their requirements, and visualize UML diagrams. (2) Application layer which contains main functionalities of the platform include NLP, generative AI, diagramming tools, and feedback systems. This layer processes user inputs, generates relevant outputs, and keeps data for analysis. (3) Data Layer: This layer consists of databases that hold user submissions, feedback, and performance metrics. It ensures smooth data retrieval and supports personalized learning paths.
The workflow begins when students enter their requirements at the interface layer. Then the application layer steps in, utilizing NLP techniques to evaluate how clear and complete those inputs are. Next, the generative AI model takes this information and write user requirements and create UML diagrams. The, feedback mechanisms offer real-time evaluations, suggesting enhancements based on established standards. Finally, the results, along with data on user engagement, are stored and analyzed in the data layer, helping to guide further development and improvements.
By making learning more engaging and targeted to individual student requirements, this study could result in better educational outcomes. Additionally, it could improve students' preparedness for real-world software engineering tasks and bridging the gap between academic learning and industry requirements.

Drawing accurate UML diagrams and writing accurate user/system requirements are two important software engineering skills that students frequently find difficult to master. Traditional education methods are lack of engagement and lack of personalized feedback. This study tackles the challenges students face in learning and applying baseline terms in software development. The project enables, through generative AI, an interactive learning of writing user/system requirements and drawing UML diagrams. The theory is that by incorporating generative AI tools into their workflow, students will better understand the underlying ideas of software development, remember them for longer, and be able to apply them in practice. Generative AI provides the following: (1) getting real-time feedback which enable students to identify and correct there mistakes immediately, (2) creating personalized learning experiences by adapts software engineering challenges and resources based on individual student performance, (3) engaging students through gamification which makes learning more attractive and interesting, (4) improve students' preparedness for real-world software engineering tasks and bridging the gap between academic learning and industry requirements, and (5) development of iterative learning processes by encourages repeated drafting, feedback, and improvement of both user/system requirements and UML diagrams.
This study proposes a multi-layered to enhance students' skills in writing user/system requirements and creating UML diagrams. (1) User interface layer which present a user-friendly and responsive interface built with front-end frameworks. It allows students to engage with the platform, input their requirements, and visualize UML diagrams. (2) Application layer which contains main functionalities of the platform include NLP, generative AI, diagramming tools, and feedback systems. This layer processes user inputs and generates relevant outputs. (3) Data Layer: This layer consists of databases that hold user submissions, feedback, and performance metrics. It ensures smooth data retrieval and supports personalized learning paths.

This study use variety of technologies to helps students enhance their skills in writing user requirements and creating UML diagrams. The first technology is Natural Language Processing frameworks like “NLTK”, and “Hugging Face’s Transformers” to analyze students’ user/system requirements structure, clarity, and completeness. The second technology involves generative AI models such as “GPT” and fine-tune it on datasets filled with examples of user/system requirements and UML diagrams and then apply prompt engineering techniques to enhance both performance. This will allow the model to generate responses and content based on prompts from students. The third technology focuses on diagramming and visualization libraries like “JointJS” for UML diagramming to enable real-time rendering of UML diagrams. The tool will transform textual descriptions into visual representations, making it easier to grasp relationships and processes. The fourth technology is Feedback Mechanisms to automatically evaluate clarity, standards, and completeness of students' work and offer a real-time feedback and suggestions for improvement based on common pitfalls identified from previous students' work. The AI models and applications will be set up on cloud servers such as “Google Cloud”, ensuring they are always available and accessible to students. Additionally, the user interface design will be features with drag-and-drop functionality and interactive feedback to make the experience even better. Student submissions, feedback, and progress data will be saved in databases like “MongoDB” allows for easy access and analysis, supporting personalized learning paths and helping us continuously improve the system. To really understand market demand, we are planning to send out surveys to educators, targeting the 90% of personalized learning. To evaluate the tool effectiveness, this study will track student performance, aiming for a 20% boost in post-assessments. It will also consider user satisfaction, targeting a 75%, and check in on faculty-reported time savings, aiming for a 50% reduction.

None

The proposed project offers an inventive solution to the hurdles students encounter when writing user/system requirements and creating UML diagrams in several significant ways: (1) Integration of Advanced technology by mix of NLP, generative AI models, and diagramming tools, this project builds a comprehensive educational platform which enables automated analysis and feedback, giving students real-time support tailored to their unique needs. (2) Personalized Learning Experience by adaptive learning pathways that respond to student performance and cater to different skill levels. The platform evaluates each student's interactions and delivers personalized content and feedback, ensuring that resources meet individual learning requirements. This customization not only enhances student engagement but also supports a variety of learning styles. (3) Engaging and Visual Learning by integrating diagramming and visualization tools, the project turns abstract ideas into visual formats. This strategy aids understanding, helping students’ better grasp complex relationships within user requirements and UML diagrams. (4) Incorporation of feedback mechanisms that use sentiment analysis and automatic grading showcases an innovative approach to student evaluation. By offering immediate, constructive feedback, the platform encourages students to reflect on their writing and make improvements right away. (5) Community-driven Learning: The project fosters a collaborative learning environment where students can share insights and support each other, enhancing the overall educational experience.

Scalability and Sustainability
Strategies for Scalability and Tackling Bottlenecks:
Cloud Infrastructure: Utilize cloud service providers like “Google Cloud” to ensure that platform can flexibly scale resources according to user demand.
Microservices Architecture: Build the system with a microservices architecture that breaks down functionalities into independently deployable services. This way, we can scale specific components in the future without impacting the whole system, effectively easing potential bottlenecks in processing and response times.
Load Balancing: Use load balancers to spread incoming user requests across several servers. This approach enhances application responsiveness by stopping any single server from becoming overwhelmed and supports horizontal scalability as user demand increases.
Fostering Long-Term User Engagement:
Gamification: adding fun elements like challenges that really get users excited to keep coming back to the platform. These features spark a bit of friendly competition and a sense of achievement, which encourages users to stick around and engage more often.
Regular Content Updates: expand the content (tutorials, resources, and case studies) on the platform so users always have something new to dive into.
User Feedback Loops: Set up ways to gather user feedback on a regular basis, giving them the chance to suggest new features or improvements. This kind of involvement makes users feel appreciated and more connected to the platform.

This study provide a platform which provides students with crucial skills in technical writing and UML diagramming regardless of their location and background, have fair access to educational resources. It also makes learning more engaging and targeted to individual student requirements which could result in better educational outcomes. Additionally, it could improve students' preparedness for real-world software engineering tasks and bridging the gap between academic learning and industry requirements.

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.