Design of AI-Enabled Assistive Communication Systems for Online Peer Learning

Peer-based online learning communities fundamentally rely on effective communication mechanisms to enable collaboration, sustain learner engagement, and promote meaningful knowledge exchange. However, real-world peer interaction within digital learning environments is often hindered by information overload, semantic ambiguity, unequal participation, accessibility constraints, and increased cognitive load. These challenges can degrade interaction quality, reduce collaborative efficiency, and ultimately impair learning outcomes. Furthermore, most existing online learning platforms employ static, non-adaptive communication tools—such as conventional discussion forums, chat modules or message boards—that lack contextual awareness and fail to dynamically adjust to evolving peer interaction patterns. To address these limitations, this study proposes an Assistive Intelligent Communication Framework (AICF) specifically designed for peer-based online learning ecosystems. The framework integrates situational awareness, adaptive decision-making, and real-time communication mediation to enhance the effectiveness of peer interactions. The proposed model captures communication dynamics through a formalized system representation in which learner interactions are modeled as contextual state transitions influenced by engagement levels, message relevance, response latency, and collaborative intent. An assistive intelligence layer—leveraging machine learning–based context inference and rule-guided intervention policies—monitors ongoing exchanges and dynamically introduces supportive mechanisms such as message summarization, clarification prompts, sentiment-aware moderation, turn-taking regulation, and adaptive notification filtering. By analyzing both semantic and behavioral interaction signals, the system adjusts communication interventions to minimize redundancy, reduce cognitive overload, and encourage balanced participation among peers. Extensive experimental evaluations were conducted using simulated and real-time peer interaction datasets, comparing the proposed framework against baseline non-adaptive communication systems. Performance was assessed using communication-centric metrics including interaction effectiveness, engagement index, response coherence, participation equity, and system overhead. The results demonstrate that the proposed model significantly enhances communication clarity and collaborative efficiency while maintaining acceptable computational and network resource utilization. Overall, the findings validate the effectiveness of incorporating assistive intelligence into peer-based learning environments. The proposed framework provides a scalable and context-aware foundation for next-generation intelligent communication systems that support inclusive, adaptive, and learner-centered digital education ecosystems.