Autism Spectrum Disorder (ASD) presents significant challenges in personalized learning due to heterogeneous cognitive, behavioral, and communication profiles among learners. Traditional educational systems often fail to provide adaptive and scalable individualized support. This research proposes a Deep Convolutional Neural Network (DCNN)-based adaptive chatbot framework designed to deliver personalized educational assistance for individuals with ASD. The system integrates deep learning-driven natural language understanding, contextual adaptation mechanisms, and behavior-aware response generation to enhance engagement and learning effectiveness.

The framework builds upon recent advancements in chatbot architectures and AI-based educational systems (Dhyani & Kumar, 2019; Kasthuri & Balaji, 2021). It also incorporates insights from domain-specific conversational AI systems trained on limited datasets (Kapočiute-Dzikiene, 2020) and ASD-oriented chatbot applications (Li et al., 2020). The proposed system further leverages architectural optimization strategies inspired by channel optimization techniques in deep learning-based chatbot frameworks (Alruily, 2022).

Experimental evaluation through simulated learning environments demonstrates that the proposed model improves response relevance, engagement consistency, and personalized learning alignment compared to conventional rule-based and hybrid chatbot systems. The study contributes to the development of AI-driven inclusive education systems by bridging gaps in accessibility, adaptability, and emotional intelligence in educational chatbots for ASD learners.