【作 者】：
王佑镁， 李宁宇， 尹以晴， 柳晨晨
【关 键 词 】：
BP神经网络； 数字阅读素养； 层级分析法； 人工智能
【栏 目】：
学习环境与资源
【中文摘要】：
针对现有数字阅读素养评测中存在的概念不清和多采用主观构建评测指标等问题，有必要引入人工智能技术来改进评测方法。首先，使用层级分析法（AHP）确定数字阅读素养各维度及其指标权重，形成面向中小学生的数字阅读素养评测指标体系；其次，利用BP神经网络训练所收集的样本数据，生成能够再现专家评价体系的神经网络模型；最后，根据所开发的数字阅读素养评测系统得到数字阅读素养评测值，利用BP神经网络学习样本中关于专家指标值和评测值的映射关系，验证数字阅读素养评测指标权重的合理性，并存储该BP神经网络用于预测中小学生数字阅读素养评测值，从而形成客观合理的数字阅读素养评测体系，为我国中小学生数字阅读评测提供模型、工具和技术系统，也为相关领域评价体系建构提供一种方法。
【英文摘要】：
Given the problems of unclear concepts and subjective construction of evaluation indicators in the existing digital reading literacy evaluation, it is necessary to introduce artificial intelligence technology to improve the evaluation method. Firstly, the analytic hierarchy process(AHP) is used to determine the dimensions of digital reading literacy and their index weights, so as to form a digital reading literacy evaluation index system for primary and secondary school students. Secondly, the sample data collected by BP neural network training is used to generate a neural network model that can reproduce the expert evaluation system. Thirdly, according to the developed digital reading literacy evaluation system, the digital reading literacy evaluation value is obtained, the mapping relationship between expert index value and evaluation value in BP neural network learning sample is used to verify the rationality of the weight of digital reading literacy evaluation index, and the BP neural network is stored to predict the digital reading literacy evaluation value of primary and secondary school students, so as to form an objective and reasonable digital reading literacy evaluation system to provide models, tools and technical systems for digital reading evaluation of primary and secondary school students in China, and also provide a method for the construction of evaluation system in related fields.