Scholarlink

<p>Evaluating exhibition communication effectiveness is essential for understanding visitor engagement, comprehension and experimental outcomes; however, traditional assessment approaches predominantly rely on subjective manual observations, limiting scalability.to overcome these limitations, the research proposes an artificial intelligence (AI)-based model, termed the Big Bang-Big Crunch driven Convolutional Refined Memory Neural Network (BBBC-CRMNN), for quantitatively assessing exhibition communication effectiveness using visitor behavioral data. Prior to model training, the dataset undergoes Z-score normalization to standardize feature scales, and median filtering to reduce noise in behavioral signals such as dwell time and movement patterns. Feature extraction is performed using a convolutional neural network (CNN) to capture spatial and interaction-related characteristics, while a refined long-short term memory (RLSTM) network models temporal engagement dynamic. The BB-BC algorithm is employed to optimize feature relevance and model hyperparameters. Experiments were conducted on an exhibition communication effectiveness dataset comprising 2100 samples with 18 behavioral and interaction attributes through Python platform. Performance evaluation and comparative analysis demonstrate that the proposed BBBC-CRMNN achieves accuracy (95.62%), precision (93.2%), recall (94.0%), F1-score (93.6%) and an AUC (96%), outperforming existing DL approaches. The results indicate that the proposed framework provides a robust and scalable solution for evaluating exhibition communication effectiveness and supporting data-driven exhibit design optimization.</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

Research on evaluating exhibition communication effectiveness using AI and neural networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月25日

查看全文
  • Xuan Wu

摘要

Evaluating exhibition communication effectiveness is essential for understanding visitor engagement, comprehension and experimental outcomes; however, traditional assessment approaches predominantly rely on subjective manual observations, limiting scalability.to overcome these limitations, the research proposes an artificial intelligence (AI)-based model, termed the Big Bang-Big Crunch driven Convolutional Refined Memory Neural Network (BBBC-CRMNN), for quantitatively assessing exhibition communication effectiveness using visitor behavioral data. Prior to model training, the dataset undergoes Z-score normalization to standardize feature scales, and median filtering to reduce noise in behavioral signals such as dwell time and movement patterns. Feature extraction is performed using a convolutional neural network (CNN) to capture spatial and interaction-related characteristics, while a refined long-short term memory (RLSTM) network models temporal engagement dynamic. The BB-BC algorithm is employed to optimize feature relevance and model hyperparameters. Experiments were conducted on an exhibition communication effectiveness dataset comprising 2100 samples with 18 behavioral and interaction attributes through Python platform. Performance evaluation and comparative analysis demonstrate that the proposed BBBC-CRMNN achieves accuracy (95.62%), precision (93.2%), recall (94.0%), F1-score (93.6%) and an AUC (96%), outperforming existing DL approaches. The results indicate that the proposed framework provides a robust and scalable solution for evaluating exhibition communication effectiveness and supporting data-driven exhibit design optimization.