Multi-source heterogeneous blockchain data quality assessment model for enterprise business activities.

Area of Science:

• Computer Science
• Information Systems
• Data Science

Background:

• Enterprise blockchain adoption is growing, but existing evaluation methods are insufficient.
• Challenges include heterogeneous data sources, inconsistent representations, and information ambiguity.
• Accurate assessment of information consistency, credibility, and value is crucial for business activities.

Purpose of the Study:

• To propose a multi-source heterogeneous blockchain data quality evaluation model for enterprise business activities.
• To address the limitations of current methods in evaluating blockchain data quality.
• To enable efficient assessment of information consistency, credibility, and value.

Main Methods:

• Developed a Representation learning for fusing entity category information (CEKGRL) model for entity representation and consistency assessment.
• Introduced a trustworthiness characterization method considering information sources, comments, and content.
• Constructed a blockchain quality assessment model integrating trustworthiness for value assessment.

Main Results:

• The proposed model significantly improves the assessment of inter-block consistency.
• Demonstrates superior performance in evaluating intra-block activity information trustworthiness.
• Achieves better assessment of the overall value of blockchain business activity information compared to existing methods.

Conclusions:

• The developed model effectively addresses the challenges of multi-source heterogeneous blockchain data quality.
• It provides a comprehensive framework for evaluating consistency, trustworthiness, and value in enterprise blockchain applications.
• The model offers a significant advancement over current evaluation techniques for business activities on the blockchain.