Research On Multi-source Heterogeneous Blockchain Data Quality Assessment Model For Enterprise Business Activities

Blockchain is a new distributed technology that generates,stores,manipulates and verifies data through a block chain structure,consensus algorithm and smart contract.It can realize value communication between nodes with untrusted relationships without relying on third-party trusted institutions.In the existing enterprise blockchain system,the information quality is uneven,and the information of business activities mostly comes from different fields and different organizations,which makes the information representation method ambiguous.At the same time,the accuracy and value of the data in the block cannot be guaranteed due to the limitation of the company’s own credibility and environmental constraints.Therefore,in the process of block establishment,there will be problems such as information inconsistency between blocks,content credibility,value and other indicators.Traditional data evaluation methods do not take advantage of the characteristics of the whole process of blockchain,which are immutable and traceable.As a result,the evaluation efficiency and accuracy are low.As a result,enterprise users and relevant regulatory authorities cannot quickly screen out the blockchain that meets their needs,nor can they establish a unified business activity analysis model.Therefore,effective evaluation of enterprise blockchain is a hot and difficult topic in the field of blockchain research.In view of the shortcomings of existing enterprise blockchain evaluation methods,this paper proposes a multi-source heterogeneous blockchain data quality evaluation model for enterprise business activities,so as to achieve efficient evaluation of business activity information consistency,credibility and value.The main research work and innovation points are as follows:(1)A method for calculating semantic similarity between blocks based on context information is proposed.Firstly,the CEKGRL model is proposed to represent the business activity information as a triplet,and the triplet information is associated with its category through the attention score.Secondly,the context information of enterprise entities is introduced to build the context structure diagram model.Finally,the similarity between blocks is calculated,and the average similarity of each block is used as the consistency evaluation result of blockchain.(2)Introduce information credibility characterization methods based on info rmation source,evaluation and content features.Information source,information evaluation and credibility of information content are used to comprehensively measure the accuracy of business activity information in blockchain,and then obtain the information accuracy evaluation results of blockchain.(3)Put forward the blockchain data quality assessment model.Firstly,the value of data in blockchain is evaluated based on the information quantity theory,and the value of blockchain data is measured by the amount of information,so as to realize the value evaluation of blockchain data.Secondly,the consistency,accuracy and value evaluation results of the blockchain are integrated,and different weights are given according to the importance of the blockchain data quality evaluation model.(4)Carry out relevant experimental verification on the open data set,and compare the experiment with the current data quality assessment method.Experimental results show that the data quality evaluation model proposed in this paper has higher evaluation efficiency.