Are Key-Foreign Key Joins Safe to Avoid when Learning High-Capacity Classifiers?
Summary: Revisits “safe” avoidance of key–foreign-key joins for decision trees, nonlinear SVMs, and ANNs. Extensive real-data and simulation results show these high-capacity models are more robust than linear classifiers, overturning prior intuition. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Vraj Shah (University of California San Diego)
- 2. Arun Kumar (University of California San Diego)
- 3. Xiaojin Zhu (University of Wisconsin)
BibTeX Citation
@article{shah_vldb18,
title = {{Are Key-Foreign Key Joins Safe to Avoid when Learning High-Capacity Classifiers?}},
author = {Shah, Vraj and Kumar, Arun and Zhu, Xiaojin},
journal = {PVLDB},
series = {{VLDB} '18},
volume = {11},
number = {3},
pages = {366--379},
doi = {10.14778/3157794.3157804},
url = {https://doi.org/10.14778/3157794.3157804},
year = {2018}
}
Incoming Citations (Sorted by Pagerank)
Showing 9 of 9 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,121 | ARDA: Automatic Relational Data Augmentation for Machine Learning | 2020 | VLDB | 0.00012093059 |
| 2,994 | Data Acquisition for Improving Machine Learning Models | 2021 | VLDB | 7.8762343e-05 |
| 3,589 | Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins | 2022 | VLDB | 7.2812353e-05 |
| 4,735 | Leva: Boosting Machine Learning Performance with Relational Embedding Data Augmentation | 2022 | SIGMOD | 6.5315782e-05 |
| 4,966 | Rotom: A Meta-Learned Data Augmentation Framework for Entity Matching, Data Cleaning, Text Classification, and Beyond | 2021 | SIGMOD | 6.4225454e-05 |
| 5,396 | Putting Things into Context: Rich Explanations for Query Answers using Join Graphs | 2021 | SIGMOD | 6.2329373e-05 |
| 7,112 | Coresets over Multiple Tables for Feature-rich and Data-efficient Machine Learning | 2023 | VLDB | 5.6990782e-05 |
| 8,842 | Data Acquisition for Improving Model Confidence | 2024 | SIGMOD | 5.3589295e-05 |
| 11,262 | Enriching Relations with Additional Attributes for ER | 2024 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 14 of 14 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 | 9,700 | Quantifying the Loss of Acyclic Join Dependencies | 2023 | PODS |
| 2 | 2,147 | Nearest Neighbor Classifiers over Incomplete Information: From Certain Answers to Certain Predictions | 2021 | VLDB |
| 3 | 7,112 | Coresets over Multiple Tables for Feature-rich and Data-efficient Machine Learning | 2023 | VLDB |
| 4 | 5,340 | Machine Learning for Databases | 2021 | VLDB |
| 5 | 9,962 | Structure-Aware Machine Learning over Multi-Relational Databases | 2021 | SIGMOD |
| 6 | 5,804 | A Relational Framework for Classifier Engineering | 2017 | PODS |
| 7 | 10,186 | Accelerating Approximate Analytical Join Queries over Unstructured Data with Statistical Guarantees | 2026 | SIGMOD |
| 8 | 12,885 | Decision Tables: Scalable Classification Exploring RDBMS Capabilities | 2000 | VLDB |
| 9 | 715 | Learning Generalized Linear Models Over Normalized Data | 2015 | SIGMOD |
| 10 | 764 | To Join or Not to Join? Thinking Twice about Joins before Feature Selection | 2016 | SIGMOD |