Enhancing Mathematics Learning through AI Chatbots and Scaffolding: A Systematic Literature Review Hasanah Sulistiyah Ningsih (1*), Hepsi Nindiasari (2), Maman Fathurrohman (3), Novaliyosi (4)
1) Mathematics Education Study Program, Universitas Sultan Ageng Tirtayasa
Jl. Raya Jakarta Km 4, Pakupatan, Serang 42124, Banten, Indonesia
*hasanahsn50[at]gmail.com
2) Mathematics Education Study Program, Universitas Sultan Ageng Tirtayasa
Jl. Raya Jakarta Km 4, Pakupatan, Serang 42124, Banten, Indonesia
3) Mathematics Education Study Program, Universitas Sultan Ageng Tirtayasa
Jl. Raya Jakarta Km 4, Pakupatan, Serang 42124, Banten, Indonesia
4)Mathematics Education Study Program, Universitas Sultan Ageng Tirtayasa
Jl. Raya Jakarta Km 4, Pakupatan, Serang 42124, Banten, Indonesia
Abstract
The rapid advancement of Artificial Intelligence (AI), particularly in the form of chatbot technologies, has created new opportunities for supporting mathematics learning. However, its pedagogical application, especially when integrated with scaffolding approaches, has not been extensively explored. Therefore, this study aims to systematically review the use of AI chatbots in mathematics learning, examine the role of scaffolding, and evaluate the effectiveness of their integration.
This study employs a Systematic Literature Review (SLR) method following PRISMA guidelines. Relevant articles were collected from several academic databases based on predefined inclusion criteria, focusing on recent studies related to AI chatbots, scaffolding, and mathematics education.
The findings reveal that AI chatbots are widely utilized as virtual tutors that provide real-time interaction, immediate feedback, and support for self-directed learning. Meanwhile, scaffolding is implemented through step-by-step guidance, prompting strategies, and feedback mechanisms that help students develop a deeper understanding of mathematical concepts. The integration of AI chatbots and scaffolding generally demonstrates positive effects on students^ learning outcomes, motivation, and engagement. However, most studies indicate that scaffolding within chatbot systems is not yet designed in an adaptive and systematic manner aligned with students^ learning needs.
These findings suggest that there is still a gap in the development of AI chatbot systems that integrate pedagogically grounded scaffolding. Therefore, future research should focus on designing and evaluating AI-based learning environments that incorporate adaptive scaffolding to support students^ problem-solving skills in mathematics learning.
Keywords: AI Chatbots- Scaffolding- Mathematics Learning- Systematic Literature Review- Problem-Solving Skills
