Rock: Cleaning Data with both ML and Logic Rules
Summary: Rock embeds ML models as predicates into logic rules, enabling joint rule-based deduction and learned inference for ER, conflict resolution, timeliness deduction and missing-value imputation across relational tables. Scalable parallel batch/incremental algorithms for rule discovery, error detection and correction with a user-friendly interface and production deployments (banks, HR). (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Zian Bao (Shenzhen University)
- 2. Binbin Bie (Shenzhen University)
- 3. Wenfei Fan (Beihang University; Shenzhen University; University of Edinburgh)
- 4. Daji Li (Shenzhen University)
- 5. Mengyun Li (Shenzhen University)
- 6. Kaiwen Lin (Shenzhen University)
- 7. Wei Lin (Shenzhen University)
- 8. Peijie Liu (Shenzhen University)
- 9. Peng Liu (Shenzhen University)
- 10. Zhicong Lv (Shenzhen University)
- 11. Mingliang Ouyang (Shenzhen University)
- 12. Chenyang Sun (Shenzhen University)
- 13. Shuai Tang (Shenzhen University)
- 14. Yaoshu Wang (Shenzhen University)
- 15. Qiyuan Wei (Shenzhen University)
- 16. Xiangqian Wu (Shenzhen University)
- 17. Min Xie (Shenzhen University)
- 18. Jing Zhang (Shenzhen University)
- 19. Runxiao Zhao (Shenzhen University)
- 20. Jie Zhu (Shenzhen University)
- 21. Yilin Zhu (Shenzhen University)
BibTeX Citation
@article{bao_vldb24,
title = {{Rock: Cleaning Data with both ML and Logic Rules}},
author = {Bao, Zian and Bie, Binbin and Fan, Wenfei and Li, Daji and Li, Mengyun and Lin, Kaiwen and Lin, Wei and Liu, Peijie and Liu, Peng and Lv, Zhicong and Ouyang, Mingliang and Sun, Chenyang and Tang, Shuai and Wang, Yaoshu and Wei, Qiyuan and Wu, Xiangqian and Xie, Min and Zhang, Jing and Zhao, Runxiao and Zhu, Jie and Zhu, Yilin},
journal = {PVLDB},
series = {{VLDB} '24},
volume = {17},
number = {12},
pages = {4373--4376},
doi = {10.14778/3685800.3685878},
url = {https://doi.org/10.14778/3685800.3685878},
year = {2024}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 10 of 10 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 33 | Consistent Query Answers in Inconsistent Databases | 1999 | PODS | 0.00049907763 |
| 112 | HoloClean: Holistic Data Repairs with Probabilistic Inference | 2017 | VLDB | 0.00032801121 |
| 1,476 | Baran: Effective Error Correction via a Unified Context Representation and Transfer Learning | 2020 | VLDB | 0.00010659277 |
| 2,297 | Raha: A Configuration-Free Error Detection System | 2019 | SIGMOD | 8.7897221e-05 |
| 6,403 | Parallel Discrepancy Detection and Incremental Detection | 2021 | VLDB | 5.8859374e-05 |
| 9,427 | Discovering Top-k Rules using Subjective and Objective Criteria | 2023 | SIGMOD | 5.271035e-05 |
| 9,647 | Rock: Cleaning Data by Embedding ML in Logic Rules | 2024 | SIGMOD | 5.2430158e-05 |
| 10,111 | Parallel Rule Discovery from Large Datasets by Sampling | 2022 | SIGMOD | 5.1347137e-05 |
| 11,424 | Splitting Tuples of Mismatched Entities | 2023 | SIGMOD | 5.093636e-05 |
| 11,434 | Learning and Deducing Temporal Orders | 2023 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 10,727 | MiniClean: A Single-Machine System for Cleaning Big Graphs | 2025 | SIGMOD |
| 2 | 647 | Discovering Data Quality Rules | 2008 | VLDB |
| 3 | 10,966 | UniClean: A Scalable Data Cleaning Solution for Mixed Errors based on Unified Cleaners and Optimized Cleaning Workflow | 2025 | VLDB |
| 4 | 11,048 | Demonstrating Matelda for Multi-Table Error Detection | 2025 | VLDB |
| 5 | 1,323 | Data Cleaning: Overview and Emerging Challenges | 2016 | SIGMOD |
| 6 | 4,890 | Descriptive and Prescriptive Data Cleaning | 2014 | SIGMOD |
| 7 | 7,880 | Learning Over Dirty Data Without Cleaning | 2020 | SIGMOD |
| 8 | 9,442 | Interactive and Deterministic Data Cleaning: A Tossed Stone Raises a Thousand Ripples | 2016 | SIGMOD |
| 9 | 9,627 | Making It Tractable to Catch Duplicates and Conflicts in Graphs | 2023 | SIGMOD |
| 10 | 9,647 | Rock: Cleaning Data by Embedding ML in Logic Rules | 2024 | SIGMOD |