Enhancing SQL Query Learning Using Chatbot with NLP: A Methodical Review

Abstract

This systematic review investigates how Artificial Intelligence (AI) and Natural Language Processing (NLP) techniques, particularly chatbot-based and text-to-SQL systems, have been applied to enhance SQL learning. Following the PRISMA framework, 26 peer-reviewed studies published between 2018 and 2025 were analyzed to identify trends, pedagogical outcomes, and research gaps. Unlike previous surveys that primarily focused on algorithmic performance, this review uniquely emphasizes the pedagogical integration of AI-driven SQL tutoring tools. A two-dimensional taxonomy was developed to categorize reviewed studies based on NLP technique type (rule-based, Seq2Seq, and Large Language Model) and application focus (educational or general-purpose). The synthesis reveals that LLM-based systems dominate recent developments, offering enhanced contextual understanding and adaptive feedback. However, challenges remain regarding scalability, error interpretation, and alignment with instructional design principles. Findings indicate that NLPpowered chatbots significantly improve learner engagement, motivation, and conceptual understanding by enabling natural language query generation and real-time feedback. Nonetheless, pedagogical grounding and long-term learning evaluation are limited. The review identifies key future directions, including pedagogically informed feedback mechanisms, scalable model design, and integration into learning management systems. Overall, this review contributes a structured synthesis and taxonomy that links technical progress with educational theory, providing a foundation for developing intelligent, theory-driven SQL tutoring systems that bridge language, logic, and learning.

Ähnliche Arbeiten