Keywords

1 Introduction

Databases are crucial in computer and information sciences, serving as the foundation for systems needing organized storage, efficient data retrieval, and analysis [6]. Teaching Database Principles is more than imparting knowledge about data manipulation and query syntax; it involves developing skills in logical structuring and strategic data organization, vital in today’s information era [2]. Traditional teaching methods, although fundamental, struggle to adapt to evolving educational needs, failing to customize instruction or fully utilize technological advancements in education. They often adopt a generic approach, neglecting the varied skills and learning paces of students.

Existing supplementary teaching systems attempt to overcome these limitations by integrating technology, but many lack the depth needed to foster critical thinking and contextual understanding. Furthermore, they often fall short in providing the personalized feedback and adaptive learning pathways essential in complex subjects like Database Principles [4]. The emergence of advanced technologies like large language models opens new possibilities.

ChatGPT-like models, with their exceptional natural language processing abilities, offer significant benefits. They enable personalized interaction, instant feedback, and content customization, aligning with personalized learning and supporting self-regulated learning [1, 3, 5]. This paper proposes a new framework for teaching Database Principles, combining a ChatGPT-like language model with traditional teaching methods to create an immersive learning environment. It aims to provide personalized learning pathways, real-time Q &A sessions, and intelligent content recommendations to enhance student understanding. Our framework’s significance extends beyond improving engagement and learning outcomes; it also seeks to provide broader access to quality educational resources and streamline Database Principles education.

The following sections discuss the framework’s concept, practical applications, potential architecture, and teaching methods, both traditional and modern. It also outlines teaching strategies, their educational impact, and concludes by suggesting evaluation methods for the framework, identifying expected strengths and challenges. Lastly, it contemplates the wider implications of AI in education and its prospects for further research.

2 System Design Concept

2.1 ChatGPT-Like Model-Based Auxiliary Teaching System

System Architecture. The architecture centers around the ChatGPT-like model, leveraging its AI capabilities (Figure.  1). This setup aims to cater to diverse student learning processes. The model excels in simulating intricate, human-like interactions, fostering an engaging and responsive learning environment. The architecture includes four main modules: a user-friendly front-end interface, a robust back-end for data processing, a comprehensive learning resource repository, and an intelligent analysis unit.The front-end acts as a portal for student interactions, collecting queries and feedback. The back-end processes student inputs through the ChatGPT-like model to generate relevant and intelligent responses. A resource library offers various educational materials, and the intelligent analysis module uses machine learning to analyze student data, enabling personalized educational experiences.

Fig. 1.
figure 1

The architecture of the proposed system.

Full size image

Natural Language Interaction Function. The model’s natural language understanding (NLU) and generation (NLG) capabilities are critical, enabling it to grasp complex student inquiries and produce coherent, accurate responses. Its technological backbone includes deep learning models trained on diverse datasets, ensuring a comprehensive grasp of database concepts and pedagogy.

Personalized Teaching Services. The system’s goal is to create an adaptive learning environment that respects individual learner differences. It uses data mining techniques and machine learning algorithms to analyze student performance, learning habits, and feedback. These insights help the system design personalized learning paths, recommend tailored resources, and adjust teaching strategies in real-time to suit different learning speeds and styles.

2.2 Blended Teaching Model

Teaching Strategy Overview. The blended teaching model combines traditional in-person teaching with online learning methods. It incorporates the ChatGPT-like auxiliary system to enhance classroom interactions and resource use, making education more dynamic and efficient.

Tiered, Phased, and Typed Teaching Methods. The design employs a tiered instruction approach, aligned with students’ existing knowledge. It progresses through phased educational content, in line with the curriculum, and offers various types of resources, such as visual aids and practical exercises, to suit different learning preferences. This tiered method ensures that materials are not one-size-fits-all but resonate with each learner’s stage and style.

Provision of Materials and Resources. The system offers a wide range of materials and resources, including foundational texts, supplementary readings, and interactive tools. These resources, dynamically curated and updated by the system, remain relevant, current, and pedagogically effective, based on ongoing teaching feedback and performance analytics.

3 Teaching Strategy Implementation

3.1 Pre-class Preparation

Video and PPT Explanations. For pre-class preparation, we use video and PPT explanations to cater to various learning styles. These materials, designed with the ChatGPT-like model, are informative and logically structured. Technologies like video editing software, content management systems, and AI-based text-to-speech engines are utilized to optimize engagement and accessibility.

Online Tests. Online tests assess students’ understanding of pre-class materials, guiding personalized teaching. These tests, administered through web-based platforms, offer immediate scoring and analytics. Machine learning algorithms analyze results, suggesting learning pathway adjustments based on individual student performance.

3.2 Application of the Auxiliary Teaching System

Personalized Exercises and Q &A. In class, our module provides tailored exercises and Q &A sessions, ensuring a suitable learning pace. The ChatGPT-like model-powered adaptive learning platform processes student inputs, evaluates their knowledge, and generates real-time questions and exercises. This dynamic adjustment employs NLU, NLG, and adaptive learning algorithms.

Real-Time Feedback and Assessment. Real-time feedback and assessment create a responsive learning environment. This functionality uses real-time analytics and educational data mining to process student inputs, offering instant feedback and navigating learning challenges.

3.3 Post-class Practice and Extension

Individualized Homework and Projects. Post-class, the system continues personalized learning with homework and projects tailored to each student’s path. This approach reinforces classroom learning through practical application. Project management tools and educational platforms help track progress and adapt project difficulty and scope to student abilities. AI assists in aligning project briefs with learning objectives.

Group Discussions and Peer Learning. Group discussions and peer learning are encouraged for collaborative understanding.

Collaboration technologies include online forums, chat rooms, and peer review systems within the teaching system. AI may guide discussions, stimulate deeper inquiries, and ensure a constructive, inclusive learning atmosphere.

4 Effectiveness Assessment and Analysis

We aim to thoroughly evaluate the auxiliary teaching system, focusing on its impact on learning. Our approach considers various dimensions of educational outcomes to gauge the system’s real-world effectiveness. We plan to use advanced data analytics, statistical software, and educational technology assessment tools for precise and comprehensive evaluation.

4.1 Evaluation Methodologies

Quantitative Performance Analytics. We may use quantitative metrics like grades, test scores, and usage statistics for evaluation. Learning analytics platforms could aggregate and analyze this data, offering a data-driven perspective on the system’s influence on student performance.

Student Surveys and Feedback. Custom-designed surveys and feedback mechanisms could gather subjective data on student satisfaction, engagement, and perceived learning gains. These surveys might use advanced scaling and branching techniques to capture detailed student responses.

Comparative Studies. Comparative studies could contrast classes using the auxiliary system with traditional methods. These studies would use experimental design techniques to ensure fair comparisons and attribute results confidently to the system’s impact.

4.2 Analysis of Course Application Effects

We plan to analyze the system’s effects on student engagement, understanding of Database Principles, and skill acquisition, potentially involving:

  • Monitoring time-on-task and interaction metrics for engagement.

  • Administering pre- and post-tests for conceptual understanding.

  • Evaluating portfolios and projects for skill development.

5 Future Directions

Future development of the system includes:

  • Enhanced Personalization: Refining adaptive algorithms and incorporating real-time feedback for a more tailored learning experience.

  • Collaborative Learning Tools: Introducing tools for group projects and discussions to foster community and enhance learning outcomes.

  • Continuous Assessment and Feedback: Implementing mechanisms for ongoing insights into student progress, guiding immediate improvements in learning.

The future of the system involves continuous improvement, innovation, and adaptability, aiming to provide an exceptional educational experience.

6 Conclusion

This paper presented a novel auxiliary teaching system for Database Principles, integrating a ChatGPT-like model to address educational challenges. It detailed the system’s architecture, strategies, and evaluation methods, showing potential for enhanced learning outcomes and efficiency. Despite limitations and areas for further research, the findings indicate significant potential in applying AI to elevate educational quality. Future work will explore deeper AI integration in education and further refine personalized teaching systems.