Research on Data Asset Value Evaluation Method Based on Integrated Machine Learning
This paper aims to seek a more accurate and general data asset value evaluation method to help stand-ardize and improve the current data transaction pricing system and promote the expansion of data transaction mar-ket scale and the transaction and circulation of data elements.Taking the detailed information of data sets and data products in Shanghai Data Exchange,Zhejiang Big Data Trading platform,Soeay Network and Deyang Data Com-ponent trading platform as samples and transaction prices,this paper builds a data asset value evaluation model that integrates the sample data of multiple big data trading websites,and compares and analyzes the evaluation effects of regression tree model and random forest regression model.The accuracy of the training set and the test set of the regression tree model were 77.91％and 72.86％,and the accuracy of the training set and the test set of the random forest regression model were 78.98％and 79.26％.The results show that the integration algorithm can improve the problem that the regression tree model is easy to overfit,and the random forest regression model has a better evaluation effect on the sample data of multiple big data trading websites.
data assetmachine learningregression treerandom forests
NETL
NSTL
万方数据
