多源异构区块链数据质量评估模型

Quality Assessment Model for Multi-source Heterogeneous Blockchain Data

区块链技术在企业经营方面的应用越来越广泛。针对经营活动中数据常有多种来源,区块与区块之间对同一信息的描述方式不同,导致信息表征不一致,这使得企业经营活动信息的一致性、可信度以及价值难以评估,基于此,论文提出一种面向企业经营活动的多源异构区块链数据质量评估方法。该方法首先提出基于CEKGRL模型的实体信息表示方法,对区块实体信息进行三元组表示,并关联其所属活动信息类别,通过上下文信息进行相似度计算,实现区块链信息一致性评估;其次,对于区块链经营活动信息的可信度评估,该方法融合了信息源、信息评价的可信度表征结果;最后,在信息可信度表征基础上,引入价值评估方法对区块链中经营活动信息的总价值进行评估,并构建区块链质量评估模型。实验结果表明,该方法在评估块间一致性、块内信息可信度及价值等方面对比现有方法分别提升了8%,16.2%和6.8%。

The application of blockchain technology in business operation is becoming more and more widespread.Based on this,this paper proposes a multi-source heterogeneous blockchain data quality assessment method for enterprise business activi⁃ties,which is difficult to assess the consistency,credibility and value of enterprise business activities.The method firstly proposes an entity information representation method based on the CEKGRL model,which represents the block entity information in a triad and associates the activity information category it belongs to,and calculates the similarity through contextual information to achieve blockchain information consistency assessment.Secondly,for the trustworthiness assessment of blockchain business activity informa⁃tion,the method integrates the trustworthiness characterization results of information sources and information evaluation.Finally,on the basis of information credibility characterization,the value assessment method is introduced to evaluate the total value of busi⁃ness activity information in the blockchain and construct a blockchain quality assessment model.The experimental results show that the method improves 8%,16.2%and 6.8%in assessing inter-block consistency,intra-block information credibility and value,re⁃spectively,compared with the existing methods.