A data driven learning approach for the assessment of data quality

Abstract

Background Data quality assessment is important but complex and task dependent. Identifying suitable measurement methods and reference ranges for assessing their results is challenging. Manually inspecting the measurement results and current data driven approaches for learning which results indicate data quality issues have considerable limitations, e.g. to identify task dependent thresholds for measurement results that indicate data quality issues. Objectives To explore the applicability and potential benefits of a data driven approach to learn task dependent knowledge about suitable measurement methods and assessment of their results. Such knowledge could be useful for others to determine whether a local data stock is suitable for a given task. Methods We started by creating artificial data with previously defined data quality issues and applied a set of generic measurement methods on this data (e.g. a method to count the number of values in a certain variable or the mean value of the values). We trained decision trees on exported measurement methods’ results and corresponding outcome data (data that indicated the data’s suitability for a use case). For evaluation, we derived rules for potential measurement methods and reference values from the decision trees and compared these regarding their coverage of the true data quality issues artificially created in the dataset. Three researchers independently derived these rules. One with knowledge about present data quality issues and two without. Results Our self-trained decision trees were able to indicate rules for 12 of 19 previously defined data quality issues. Learned knowledge about measurement methods and their assessment was complementary to manual interpretation of measurement methods’ results. Conclusions Our data driven approach derives sensible knowledge for task dependent data quality assessment and complements other current approaches. Based on labeled measurement methods’ results as training data, our approach successfully suggested applicable rules for checking data quality characteristics that determine whether a dataset is suitable for a given task.

Targeted Data Quality Analysis for a Clinical Decision Support System for SIRS Detection in Critically Ill Pediatric Patients

“…This endorses an assumption from testing the data-driven approach with artificial data: the data-driven approach is valuable in particular for cases where the data quality issue is nothing obviously odd (like outlier values) or a perfect cause of failure (like a CDSS always failing with error). 25 Since decision trees basically just perform a lot of statistical tests (ANOVA) on all feature variables, they are a handy method to get a grip on hidden correlations between MM-results and CDSS performance. The approach's current limitation is the uncertain reliability of the derived information from the decision trees.…”

“…This is why we applied a data-driven learning approach for data quality assessment. 25 This approach used the MMs' results as features to train a decision tree and the CDSS performance data as labels. Each row of the features in the training dataset consisted of all MMs' result values for the respective patient and day.…”

Exploring the Design of Teaching Mental Health of College Students Based on Data-Driven Learning

“…e basic concept of invitational education is that schools should respect the value and potential of each individual in all aspects of the educational process, and fully mobilize all school personnel and material conditions [16]. Tute et al creatively proposed a "six-in-one" psychological support network model for students in the social context of the information age, according to the laws and characteristics of students' psychological development, with the purpose of "helping people to help themselves and growing up with mutual help," with students as the main body, with the network as the carrier, and with environmental construction as the basis [17]. "Mens et al proposed a comprehensive model of psychological counseling, the core idea of which is to highlight the status and role of counseling teachers in school education and to specify the role of counseling teachers [18].…”