您当前的位置:首页 > 论文详情

科普类微博信息质量评价指标体系构建与实证研究

请选择邀稿期刊:

Construction of an Evaluation Index System for the Quality of Scientific Popularization Posts on Weibo and Empirical Study

摘要: 【目的/意义】构建科普类微博信息质量评价指标体系并训练RBF神经网络进行评价模拟,为提升科普类微博信息质量提供参考。【方法/过程】基于现有研究成果与用户问卷调研,构建评价指标体系。根据指标体系获取评分数据进行实证,通过CRITIC-TOPSIS综合评价对25位科普类博主进行评分并利用评分数据训练RBF神经网络。【结果/结论】构建了包含5个一级指标与18个二级指标的评价指标体系,训练的神经网络可以模拟复杂多维指标体系下的信息质量评价,直接输出评价结果,精确率达到96%。为科普类微博信息质量评价提供了可参考的指标体系与准确、易用的评价方法。

Abstract: [Purpose/significance] This study aims to construct an evaluation index system for the quality of scientific popularization posts on Weibo and to simulate the evaluation using an RBF neural network. The goal is to provide a reference for improving the quality of scientific content on Weibo. [Method/process] Based on existing research and a user survey, an evaluation index system was developed. The empirical data was collected and scored according to this system. The CRITIC-TOPSIS method was used to comprehensively evaluate 25 popular science bloggers, and the scores were used to train the RBF neural network. [Results/conclusion] An evaluation index system comprising 5 primary indicators and 18 secondary indicators was constructed. The trained neural network can simulate the quality assessment of information under a complex, multi-dimensional index system and directly output evaluation results with an accuracy of 96%. This provides a useful index system and an accurate, easy-to-use evaluation method for assessing the quality of popular science posts on Weibo.

版本历史

[V1] 2024-09-19 16:28:39 PSSXiv:202409.01318V1 下载全文
点击下载全文
在线阅读
许可声明
metrics指标
  •  点击量33
  •  下载量8
  • 评论量 0
评论
分享
邀请专家评阅
收藏