Development and Evaluation of a Conversational Natural Language Processing Model for Technology-Enhanced Biology Learning in Secondary Schools

Abstract Background The integration of digital technology into education has significantly transformed how students’ learning complex scientific subjects such as biology. Objective This study developed and evaluated a Conversational Natural Language Processing Model (CNLPM) designed to enhance secondary school biology learning in Niger State, Nigeria. Methology: The model was created via the ADDIE instructional design framework and evaluated via a quasi-experimental research design to determine its effects on students’ achievement, retention, and interest. A developmental research approach was adopted, combining the formative phases of analysis, design, development, implementation, and evaluation with an experimental validation stage. Six secondary schools were selected via a multistage sampling technique, yielding an intact sample of 261 senior secondary II biology students (146 males and 115 females). Three schools were assigned to the experimental group, which received instruction via the CNLPM, whereas the remaining three formed the control group and were taught via the conventional lecture method. Result The findings revealed that students exposed to the CNLPM significantly outperformed their counterparts in the control group (mean gains of 49.46 vs. 28.14). The model also contributed to improved retention and heightened interest in biology learning. These results indicate that conversational AI tools can meaningfully complement traditional instructional strategies. This study provides evidence supporting the adoption of technology-enhanced learning approaches to improve biology education in secondary schools. Conclusion Although the study was conducted in Niger State, Nigeria, the challenges addressed—limited access to interactive learning resources, variability in teacher expertise, and low student engagement—are common across many secondary education systems worldwide. Consequently, the findings offer transferable insights into the potential of conversational AI tools for enhancing science education globally, providing scalable strategies to improve learning outcomes in diverse educational contexts. Unique Contribution: This study makes a unique contribution by empirically demonstrating how an offline-capable conversational AI model, designed for resource-constrained secondary school contexts, can simultaneously enhance conceptual understanding, learner motivation, and equitable access to biology education, while directly advancing the goals of United Nations Sustainable Development Goal 4 (Quality Education) through inclusive and scalable AI-supported instruction