Page Content
Paper Accepted for SSDBM 2019
The paper
"REDS: Estimating the Performance
of Error Detection Strategies Based on Dirtiness
Profiles"
by Mohammad Mahdavi and Ziawasch Abedjan was accepted for SSDBM 2019.
Here is the abstract:
Datasets usually suffer from various data
quality problems or data errors. At the same time, there are various
error detection strategies to detect different kinds of data errors.
To effectively detect the data errors, the user has to deploy and test
multiple error detection strategies. However, evaluating each error
detection strategy on a new dataset requires tedious manual evaluation
efforts. Therefore, estimating the performance of each strategy
upfront is desirable for a more effective strategy
selection.
In this paper, we propose a new approach to estimate the performance of error detection strategies. Our intuition is that error detection strategies will perform similarly on similarly dirty datasets. We introduce the novel concept of dirtiness profiles, which make datasets comparable with respect to their dirtiness.
Our experiments show that our system REDS accurately estimates the performance of error detection strategies and, solely based on automatically extracted features, outperforms the semi-supervised baseline.
Zusatzinformationen / Extras
Quick Access:
Schnellnavigation zur Seite über Nummerneingabe
Auxiliary Functions
Copyright TU Berlin 2008