Resumen
With the rapid growth of customer data in financial institutions, such as trusts, issues of data quality have become increasingly prominent. The main challenge lies in constructing an effective evaluation method that ensures accurate and efficient assessment of customer data quality when dealing with massive customer data. In this paper, we construct a data quality evaluation index system based on the analytic hierarchy process through a comprehensive investigation of existing research on data quality. Then, redundant features are filtered based on the Shapley value, and the multiple linear regression model is employed to adjust the weight of different indices. Finally, a case study of the customer and institution information of a trust institution is conducted. The results demonstrate that the utilization of completeness, accuracy, timeliness, consistency, uniqueness, and compliance to establish a quality evaluation index system proves instrumental in conducting extensive and in-depth research on data quality measurement dimensions. Additionally, the data quality evaluation approach based on multiple linear regression facilitates the batch scoring of data, and the incorporation of the Shapley value facilitates the elimination of invalid features. This enables the intelligent evaluation of large-scale data quality for financial data.