Entity Matching with AUC-Based Fairness
Bibliographic record
Abstract
The research on fair machine learning (ML) has been growing due to the high demand for unbiased and fair ML models for objective decision-making. Most of this research has been focused on training and tuning the ML model, and less effort has been made to study biases in the processes that clean and prepare data for these models. This paper studies fairness in entity matching (EM), a.k.a. record matching and entity resolution, a primary task in a data cleaning pipeline that can significantly impact ML models’ performance. We introduce a new metric for measuring bias in EM based on Area Under the Curve (AUC) and the risk of record matching between and within subpopulations. We use this metric and real-world data to show biases in a state-of-the-art EM technique. We introduce a debiasing algorithm based on data augmentation (DA) to mitigate bias and conduct experiments to show the algorithm’s effectiveness.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.027 | 0.136 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.005 | 0.009 |
| Open science | 0.004 | 0.007 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".