Baran: Effective Error Correction via a Unified Context Representation and Transfer Learning
Summary: Baran introduces a unified context representation and transfer learning to fuse multiple error-corrector models for data repair. Modeling full context—value, tuple co-occurrences, and attribute type—yields richer candidates and higher precision, with Wikipedia pretraining boosting recall and needing ~20 labeled tuples. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Mohammad Mahdavi (Technical University of Berlin)
- 2. Ziawasch Abedjan (Technical University of Berlin)
BibTeX Citation
@article{mahdavi_vldb20,
title = {{Baran: Effective Error Correction via a Unified Context Representation and Transfer Learning}},
author = {Mahdavi, Mohammad and Abedjan, Ziawasch},
journal = {PVLDB},
series = {{VLDB} '20},
volume = {13},
number = {11},
pages = {1948--1961},
doi = {10.14778/3407790.3407801},
url = {https://doi.org/10.14778/3407790.3407801},
year = {2020}
}Incoming Citations (Sorted by Pagerank)
Showing 34 of 34 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 19 of 19 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,827 | QFix: Diagnosing Errors through Query Histories | 2017 | SIGMOD |
| 2 | 12,337 | BART in Action: Error Generation and Empirical Evaluations of Data-Cleaning Systems | 2016 | SIGMOD |
| 3 | 11,738 | Splitting Tuples of Mismatched Entities | 2023 | SIGMOD |
| 4 | 8,328 | Robust Entity Resolution using Random Graphs | 2018 | SIGMOD |
| 5 | 2,964 | Towards Dependable Data Repairing with Fixing Rules | 2014 | SIGMOD |
| 6 | 2,960 | Auto-Detect: Data-Driven Error Detection in Tables | 2018 | SIGMOD |
| 7 | 1,805 | Raha: A Configuration-Free Error Detection System | 2019 | SIGMOD |
| 8 | 989 | Towards Certain Fixes with Editing Rules and Master Data | 2010 | VLDB |
| 9 | 5,679 | Mining an "Anti-Knowledge Base" from Wikipedia Updates with Applications to Fact Checking and Beyond | 2020 | VLDB |
| 10 | 5,954 | Semi-Supervised Data Cleaning with Raha and Baran | 2021 | CIDR |