Development of an AI-based modular system for automated assessment and adaptive learning in higher school

The growing complexity of educational demands and the expansion of digital learning environments have underscored the need for intelligent automation in teaching and assessment processes. Traditional educational platforms often lack adaptability, resulting in limited personalization and increased workload for educators. The integration of artificial intelligence (AI) into learning systems presents a promising avenue for addressing these challenges by enhancing scalability, efficiency, and individualized instruction. This study aimed to improve the quality and efficiency of the educational process by developing and implementing an AI-powered automated system for generating, verifying, and analyzing educational control tasks. The system was designed to support personalized learning, streamline assessment, and reduce the burden of routine academic activities. The proposed solution is built on a modular architecture using contemporary web technologies (React, Next.js, Firebase) in combination with the GPT model API. The system includes modules for test generation, automated answer checking, a conversational AI assistant, performance analytics, and real-time feedback. Document processing capabilities (DOCX, PDF) and seamless integration with Google Forms are also incorporated. The system’s performance was evaluated based on assessment accuracy, time savings, and usability. Implementation results indicate high effectiveness of the system in real educational scenarios. The automated evaluation module achieved an accuracy rate of 80–96 %, closely aligning with manual grading benchmarks. Additionally, the time required to prepare instructional content and assessments was reduced by 60–80 %. The user interface enabled intuitive access to system functionalities, and the adaptive features provided a personalized experience for students of varying proficiency levels. The developed system demonstrates significant potential for transforming educational practices through AI integration. It enhances personalization, reduces educator workload, and improves the consistency and objectivity of assessments. Future research will focus on expanding system functionality, including support for multimodal learning and large-scale institutional deployment.