Using Embedding Models to Improve Probabilistic Race Prediction

• URL: http://arxiv.org/abs/2604.22555v2
• Date: Tue, 28 Apr 2026 01:47:04 GMT
• Title: Using Embedding Models to Improve Probabilistic Race Prediction
• Abstract summary: Estimating racial disparity requires individual-level race data.<n>Data capture race-surname relationships only for common surnames, omitting approximately 10% of the US population.<n>We propose embedding-powered BISG (eBISG) to estimate race probabilities for names not covered in the Census.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Estimating racial disparity requires individual-level race data, which are often unavailable due to the sensitivity of collecting such information. To address this problem, many researchers utilize Bayesian Improved Surname Geocoding (BISG), which have critically relied on Census surname data. Unfortunately, these data capture race-surname relationships only for common surnames, omitting approximately 10% of the US population. We show that predictive performance degrades substantially for individuals with such omitted, uncommon surnames because standard BISG implementation relies on a uninformative generic prior in these cases. To address this limitation, we propose embedding-powered BISG (eBISG), which uses pre-trained text embeddings to represent names as dense vectors and trains neural networks on 2020 Census surname and first-name data to estimate race probabilities for names not covered in the Census. We compare five approaches: standard BISG using only surnames, BIFSG incorporating first name probabilities, surname embedding for unlisted names, surname and first name embedding combining both, and a full-name embedding trained on voter file data from Southern states that captures interactions between name components. We show that each successive eBISG approach improves race prediction, with the full-name embedding yielding the largest gains, particularly for Hispanic and Asian voters whose surnames are absent from the Census list.

Related papers

• Multicultural Name Recognition For Previously Unseen Names [65.268245109828]
  This paper attempts to improve recognition of person names, a diverse category that can grow any time someone is born or changes their name. I look at names from 103 countries to compare how well the model performs on names from different cultures. I find that a model with combined character and word input outperforms word-only models and may improve on accuracy compared to classical NER models.
  arXiv  Detail & Related papers  (2024-01-23T17:58:38Z)
• UFineBench: Towards Text-based Person Retrieval with Ultra-fine Granularity [50.91030850662369]
  Existing text-based person retrieval datasets often have relatively coarse-grained text annotations. This hinders the model to comprehend the fine-grained semantics of query texts in real scenarios. We contribute a new benchmark named textbfUFineBench for text-based person retrieval with ultra-fine granularity.
  arXiv  Detail & Related papers  (2023-12-06T11:50:14Z)
• SpellMapper: A non-autoregressive neural spellchecker for ASR customization with candidate retrieval based on n-gram mappings [76.87664008338317]
  Contextual spelling correction models are an alternative to shallow fusion to improve automatic speech recognition. We propose a novel algorithm for candidate retrieval based on misspelled n-gram mappings. Experiments on Spoken Wikipedia show 21.4% word error rate improvement compared to a baseline ASR system.
  arXiv  Detail & Related papers  (2023-06-04T10:00:12Z)
• Disambiguation of Company names via Deep Recurrent Networks [101.90357454833845]
  We propose a Siamese LSTM Network approach to extract -- via supervised learning -- an embedding of company name strings. We analyse how an Active Learning approach to prioritise the samples to be labelled leads to a more efficient overall learning pipeline.
  arXiv  Detail & Related papers  (2023-03-07T15:07:57Z)
• Estimating Racial Disparities When Race is Not Observed [3.0931877196387196]
  We introduce a new class of models that produce racial disparity estimates by using surnames as an instrumental variable for race. A validation study based on the North Carolina voter file shows that BISG+BIRDiE reduces error by up to 84% when estimating racial differences in party registration. We apply the proposed methodology to estimate racial differences in who benefits from the home mortgage interest deduction using individual-level tax data from the U.S. Internal Revenue Service.
  arXiv  Detail & Related papers  (2023-03-05T04:46:16Z)
• Race and ethnicity data for first, middle, and last names [0.0]
  We provide the largest compiled publicly available dictionaries of first, middle, and last names for imputing race and ethnicity. The dictionaries are based on the voter files of six Southern states that collect self-reported racial data upon voter registration.
  arXiv  Detail & Related papers  (2022-08-26T05:27:50Z)
• Addressing Census data problems in race imputation via fully Bayesian Improved Surname Geocoding and name supplements [0.0]
  We introduce a fully Bayesian Improved Surname Geocoding (fBISG) methodology that accounts for potential measurement error in Census counts. We supplement the Census surname data with additional data on last, first, and middle names taken from the voter files of six Southern states where self-reported race is available. Our empirical validation shows that the fBISG methodology and name supplements significantly improve the accuracy of race imputation across all racial groups, and especially for Asians.
  arXiv  Detail & Related papers  (2022-05-12T14:41:45Z)
• Rethnicity: Predicting Ethnicity from Names [0.0]
  I use the Bidirectional LSTM as the model and Florida Voter Registration as training data. Special care is given for the accuracy of minority groups, by adjusting the imbalance in the dataset.
  arXiv  Detail & Related papers  (2021-09-19T21:30:22Z)
• Multilingual Autoregressive Entity Linking [49.35994386221958]
  mGENRE is a sequence-to-sequence system for the Multilingual Entity Linking problem. For a mention in a given language, mGENRE predicts the name of the target entity left-to-right, token-by-token. We show the efficacy of our approach through extensive evaluation including experiments on three popular MEL benchmarks.
  arXiv  Detail & Related papers  (2021-03-23T13:25:55Z)
• On Leveraging Unlabeled Data for Concurrent Positive-Unlabeled Classification and Robust Generation [72.062661402124]
  We present a novel training framework to jointly target PU classification and conditional generation when exposed to extra data.<n>We prove the optimal condition of CNI-CGAN and experimentally, we conducted extensive evaluations on diverse datasets.
  arXiv  Detail & Related papers  (2020-06-14T08:27:40Z)
• Predicting Race and Ethnicity From the Sequence of Characters in a Name [0.0]
  We model the relationship between characters in a name and race and ethnicity using various techniques. A model using Long Short-Term Memory works best with out-of-sample accuracy of.85. The best-performing last-name model achieves out-of-sample accuracy of.81.
  arXiv  Detail & Related papers  (2018-05-05T20:04:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.