A Non-AI Homework Support Tool to Enhance Achievement and Interest in Science Education: BilgeCan Bot
Year: 2025
Type: article
Study design: Clinical Trial
Abstract: Homework helps students learn and develop independent study skills, but they often need extra guidance and reliable information sources, especially in science where abstract concepts can be challenging. Traditional resources may not match students’ cognitive levels and can lead to information overload. To address this, non-AI rule-based (NARB) educational chatbots can provide focused, essential information with minimal cognitive load. This study explores the effects of BilgeCan Bot, a NARB chatbot designed to help middle school students with astronomy homework, based on Cognitive Load Theory. Conducted during the 2022–2023 academic year with 52 fifth-grade students in Türkiye, the research used an explanatory sequential mixed-methods approach. Students were divided into an experimental group using BilgeCan Bot and a control group using textbooks. Data was collected through tests, interest scales, interviews, chatbot records, and teacher feedback. Results showed that students using BilgeCan Bot achieved higher scores and had greater interest in science. They described the chatbot as effective and engaging, helping them grasp difficult concepts. Overall, the findings suggest that NARB chatbots can offer targeted, reliable support for homework in science education without overwhelming students.
Authors: Taner Yılmaz, İlbilge Dökme
Institutions: Yozgat Bozok Üniversitesi, Gazi University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4414837832