摘要
[目的/意义]在数字经济背景下,数据资源通过网络平台交易实现价值转化与增值。作为一种非标准化的新兴商品,数据资源具有成本模糊、类型多样、不确定性高等典型特征,传统价值评估理论难以对其价值进行准确衡量。[方法/过程]对此,文章提出了AGA-BP神经网络的数据资源价值评估模型,该模型充分考虑了诸多影响因素与数据资源价值的非线性关系,通过自适应遗传算法(AGA)优化传统BP神经网络提升价值评估的精度,解决BP神经网络极易陷入局部最优、收敛速度较慢等问题。以此方法为基础,收集武汉东湖大数据交易中心的244条数据资源交易信息进行实证检验。[结果/结论]研究结果表明:基于AGA-BP神经网络的数据资源价值评估方法相比于GA-BP神经网络和BP神经网络性能提升明显;该方法在仿真能力、误差水平、拟合数据能力等方面表现出突出优势,具有更好的价值评估仿真效果。该方法在减少数据交易平台买卖双方交易成本,完善数据交易平台的定价机制和策略方面具有较强的指导意义。
[Purpose/significance] In the context of the digital economy,data resources realize value conversion and valueadded through online platform transaction. As a non-standardized emerging commodity,data resources have typical characteristics such as cost ambiguity,various types,and high uncertainty. It is difficult for traditional value assessment theory to accurately measure its value. [Method/process] In this regard,this study proposes a data resources value assessment model for AGA-BP neural network,which fully considers the nonlinear relationship between many influencing factors and data resources value,optimizes traditional BP neural network through adaptive genetic algorithm( AGA) to enhance the accuracy of the value assessment and solve the problem that the BP neural network is easy to fall into local optimum and the convergence speed is slow. Based on this method,244 data resources transaction information of Wuhan East Lake Big Data Trading Center was collected for empirical testing. [Result/conclusion] The results show that the data resources value assessment method based on AGA-BP neural network is more effective than GA-BP neural network and BP neural network. The method is outstanding in terms of simulation ability,error level and fitting data ability and has better value assessment simulation effects. This method has a strong guiding significance in reducing the transaction costs of buyers and sellers of data transaction platform and improving the pricing mechanism and strategy of data transaction platform.
出处
《情报理论与实践》
CSSCI
北大核心
2020年第1期135-142,共8页
Information Studies:Theory & Application
基金
国家重点研发计划项目“服务价值与文化传播评估理论与技术”(项目编号:2017YFB1400400)
北京市青年拔尖人才培育项目(项目编号:CIT&TCD20180405)
勤信英才项目(项目编号:QXTCPC201706)
北京信息科技大学研究生科技创新项目“面向交易便利化的文化类数据资源价值评估体系及智能方法研究”(项目编号:5121911045)的成果
