Validating Data Quality Actions in Scoring Processes

Abstract

Data quality has gained momentum among organizations upon the realization that poor data quality might cause failures and/or inefficiencies, thus compromising business processes and application results. However, enterprises often adopt data quality assessment and improvement methods based on practical and empirical approaches without conducting a rigorous analysis of the data quality issues and outcome of the enacted data quality improvement practices. In particular, data quality management, especially the identification of the data quality dimensions to be monitored and improved, is performed by knowledge workers on the basis of their skills and experience. Control methods are therefore designed on the basis of expected and evident quality problems; thus, these methods may not be effective in dealing with unknown and/or unexpected problems. This article aims to provide a methodology, based on fault injection, for validating the data quality actions used by organizations. We show how it is possible to check whether the adopted techniques properly monitor the real issues that may damage business processes. At this stage, we focus on scoring processes, i.e., those in which the output represents the evaluation or ranking of a specific object. We show the effectiveness of our proposal by means of a case study in the financial risk management area.

Estimating the Likelihood of Financial Behaviours Using Nearest Neighbors

“…Data quality is important for reliable results (Cappiello et al, 2018). However, this falls out of the scope of this work, which works directly with the results produced by the model rather than verifying the underlying variables.…”

A systematic literature review on data quality assessment

“…For instance, the data quality validation methodology (DQVM) has as its objective the assessment of the effects of bad data quality as well as the analysis of associated data quality actions on the results of processes, particularly scoring processes that produce as their output an evaluation that is a ranking or rating for an object. This methodology suggests a series of stages to examine the consequences of faults, injecting faults methodically throughout the process to identify various abnormal circumstances [104]. Furthermore, the total meteorological ❒ ISSN: 2302-9285 data quality (TMDQ) framework, based on the total quality management (TQM) approach developed by [115], aims to offer observers with diverse meteorological data qualities from numerous perspectives according to four quality dimensions, accuracy, consistency, completeness, and timeliness.…”

A Multi-layered Data Preparation Model for Health Information in Sudan

“…For instance, Heinrich et al introduced a metric-based approach to quantify the data quality by adding data correctness and a timeline to meet the system requirements [19]. Rather than using pre-defined requirements as recommended by Heinrich [19], Cappiello et.al introduced another approach which uses scoring processes to compare the result outputs to match the evaluation of pre-specified objects [20]. In addition, an assessment of data quality was introduced in 2018 using the Data Quality Framework (DQF) by creating quality properties and provenance of data with respect to user's experience of the quality, to make it possible to assess the data quality and track possible data errors [21].…”