Scholarlink

<p>AI-enabled smart service computing seamlessly coordinates data, devices, and users to provide IoT networks with intelligent, adaptive, and context-aware services. Interaction systems match machine decisions to human purpose, cognition, and usability. Most IoT service computing models overlook human intent awareness, cognitive state fluctuations, and interface transparency, resulting in service misunderstanding, lower trust, and restricted customisation in dynamic scenarios. To overcome these limits, this work uses Cognitive-Aware Intent-Adaptive Smart Service Computing (CAI-ASSC) to describe human intent and cognitive state in IoT service orchestration. Intent-adaptive reinforcement learning dynamically adjusts service choices based on explicit and implicit interaction signals, deep learning models, and user intent and cognitive load. Trust-preserving explainability modules increase transparency and user confidence by altering automation levels based on interaction circumstances. Intelligent living and assistive IoT solutions need adaptive interaction and user alignment. CAI-ASSC successfully reduces cognitive load by 17.5–24.6%, maintains high user satisfaction scores of 85–79 across interaction scenarios, achieves service accuracy of 94.0%, and reduces decision adaptation latency to 180–300&#xa0;ms, validating its efficacy for next-generation intelligent IoT networks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cognitive-Aware Intent-Adaptive Smart Service for explainable and trustworthy human-centric IoT networks in Digital Marketing

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月17日

查看全文
  • Yegen Ouyang,
  • Bingfeng Liu,
  • Pinghua Wu,
  • Ling Guo

摘要

AI-enabled smart service computing seamlessly coordinates data, devices, and users to provide IoT networks with intelligent, adaptive, and context-aware services. Interaction systems match machine decisions to human purpose, cognition, and usability. Most IoT service computing models overlook human intent awareness, cognitive state fluctuations, and interface transparency, resulting in service misunderstanding, lower trust, and restricted customisation in dynamic scenarios. To overcome these limits, this work uses Cognitive-Aware Intent-Adaptive Smart Service Computing (CAI-ASSC) to describe human intent and cognitive state in IoT service orchestration. Intent-adaptive reinforcement learning dynamically adjusts service choices based on explicit and implicit interaction signals, deep learning models, and user intent and cognitive load. Trust-preserving explainability modules increase transparency and user confidence by altering automation levels based on interaction circumstances. Intelligent living and assistive IoT solutions need adaptive interaction and user alignment. CAI-ASSC successfully reduces cognitive load by 17.5–24.6%, maintains high user satisfaction scores of 85–79 across interaction scenarios, achieves service accuracy of 94.0%, and reduces decision adaptation latency to 180–300 ms, validating its efficacy for next-generation intelligent IoT networks.