<b>Modeling and </b><b>O</b><b>ptimization of </b><b>L</b><b>ibrary </b><b>I</b><b>nformation </b><b>O</b><b>verload </b><b>C</b><b>ognitive </b><b>L</b><b>oad </b><b>B</b><b>ased on </b><b>M</b><b>ulti-factor </b><b>I</b><b>nteraction </b><b>M</b><b>odel and CNN-LSTM </b><b>F</b><b>usion </b><b>N</b><b>etwork</b>

With the explosive growth of library information resources driven by big data and artificial intelligence technology, users are facing increasingly serious information overload, which significantly increases cognitive load and affects learning and mental health. Therefore, this paper integrates cognitive neuroscience and deep learning, constructs a multi-factor interaction model, incorporates variables such as information amount, task complexity, user background, and time pressure into a unified framework, and uses a CNN-LSTM hybrid network to perform real-time feature extraction and fusion of 256 Hz EEG signals and behavioral data to achieve high-precision prediction of cognitive load在大数据与人工智能技术驱动下，图书馆信息资源呈爆发式增长，用户面临的信息过载问题日益严峻，该状况显著提升了认知负荷，并对学习与心理健康造成不良影响。为此，本文融合认知神经科学与深度学习技术，构建多因素交互模型，将信息总量、任务复杂度、用户背景及时间压力等变量纳入统一框架，并采用卷积长短期记忆（Convolutional Neural Network-Long Short-Term Memory, CNN-LSTM）混合网络对256 Hz脑电图（Electroencephalogram, EEG）信号与行为数据开展实时特征提取与融合，以实现认知负荷的高精度预测。

搜集汇总

背景与挑战

背景概述

该数据集聚焦于图书馆信息过载背景下的认知负荷研究，通过整合多因素交互模型和CNN-LSTM融合网络，对256Hz脑电图信号及行为数据进行分析，旨在实现高精度的认知负荷预测。数据集包含原始实验数据，适用于人工智能和认知神经科学领域的交叉研究。