Published Paper Details:

Mathematical word problems (MWPs) are a persistent challenge in education due to their dual requirement for natural language understanding and symbolic reasoning. Traditional automated solvers often provide static, symbol-based answers, limiting accessibility for learners who prefer visual or auditory approaches. Recent developments in deep learning and multi-modal educational technologies have improved solver accuracy and engagement but typically require extensive datasets or high-end hardware, making them impractical for widespread classroom use. This paper introduces the Math visualization Generator (MVG), a Python-driven system integrating natural language processing, structured reasoning, and dynamic visualization. MVG delivers multimodal explanations by generating either stepwise textual breakdowns or synchronized video-audio solutions, adapting to diverse learner needs. Tested across algebraic, geometric, and arithmetic MWPs, MVG achieves over 90% accuracy and significantly increases engagement among learners. Its lightweight, flexible design positions MVG as an effective tool bridging the gap between advanced computational reasoning and practical, accessible math instruction for classrooms and independent study