Research Of Pricing Mechainism Based On Data Quality

In this paper,the existing data pricing algorithm does not consider the impact of data quality,there is no data pricing method based on data quality evaluation,the existing data tuple pricing method takes a long time,and the query-based pricing mechanism is not flexible enough.A data quality evaluation method for data sets,and with data quality as the core,a dynamic data pricing method based on Stackelberg game is proposed.Finally,experiments prove the rationality of the theory proposed in this paper.The specific work of this paper and the results obtained are as follows:(1)A data quality evaluation method based on Shapley value is proposed.This method is based on the characteristics that the isolated forest algorithm can still maintain good performance on large-scale data sets,can quickly divide the data set into data subsets of different quality levels,and propose a distance-based quality scoring mechanism based on the divided data set.Although the distance-based data quality evaluation method can evaluate the inherent quality of the data set to a certain extent,in the combination of multiple data sets,it is impossible to determine the contribution of a certain data set to the entire model.Therefore,this article first evaluates the quality of the data set to obtain the quality information of the data set;then by combining the data set,the data set is trained to obtain the model accuracy on this data set,and then through the Shapley value method,Determine the contribution of a single data set to the entire model to evaluate the quality of the data set.(2)Propose a dynamic pricing method of Stackelberg game with data quality as the core.According to the Shapley value calculation method proposed in this paper,the data quality value of the data set is calculated,the data quality is used as a factor of the game,combined with the Stackelberg game model,the data provider is the first mover of the game,and the data buyer is the last of the game.To build a dynamic data pricing model.The modified model can dynamically price each data set requested by the data buyer according to the user's request.Through the Stackelberg model,the optimal quality division strategy and pricing strategy for each data set are calculated,which can be used as the reference price for both buyers and sellers of data.(3)Experimental verification of the pricing model.The program proposed in this paper is verified by experiments,a simulation system is written,and the data set collected by the crawler and the user's bid are used as the input of the model,and the pricing model is tested.The experimental results show that the data quality evaluation method proposed in this paper can correctly evaluate the quality level of the data set.Compared with the traditional query-based pricing method,the game pricing method proposed in this paper has higher pricing accuracy and better pricing accuracy.Low time complexity and more flexible pricing.