Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/28873
Title: | Bayesboost: Identifying and Handling Bias Using Synthetic Data Generators |
Authors: | Draghi, B Wang, Z Myles, P Tucker, A |
Keywords: | synthetic data generators;data bias;over-sampling;Bayesian network |
Issue Date: | 1-Sep-2021 |
Publisher: | ML Research Press |
Citation: | Draghi, B. et al. (2021) 'Bayesboost: Identifying and Handling Bias Using Synthetic Data Generators', Proceedings of Machine Learning Research, 154, pp. 49-62. Available at: https://proceedings.mlr.press/v154/draghi21a.html (Accessed: 28 March 2023). |
Abstract: | Advanced synthetic data generators can model sensitive personal datasets by creating simulated samples of data with realistic correlation structures and distributions, but with a greatly reduced risk of identifying individuals. This has huge potential in medicine where sensitive patient data can be simulated and shared, enabling the development and robust validation of new AI technologies for diagnosis and disease management. However, even when massive ground truth datasets are available (such as UK-NHS databases which contain patient records in the order of millions) there is a high risk that biases still exist which are carried over to the data generators. For example, certain cohorts of patients may be under-represented due to cultural sensitivities amongst some communities, or due to institutionalised procedures in data collection. The under-representation of groups is one of the forms in which bias can manifest itself in machine learning, and it is the one we investigate in this work.These factors may also lead to structurally missing data or incorrect correlations and distributions which will be mirrored in the synthetic data generated from biased ground truth datasets. In this paper, we explore methods to improve synthetic data generators by using probabilistic methods to firstly identify the difficult to predict data samples in ground truth data, and then to boost these types of data when generating synthetic samples. The paper explores attempts to create synthetic data that contain more realistic distributions and that lead to predictive models with better performance. |
Description: | Paper presented at the Third International Workshop on Learning with Imbalanced Domains: Theory and Applications (LIDTA 2021), online, 17 September 2021. |
URI: | https://bura.brunel.ac.uk/handle/2438/28873 |
ISSN: | 2640-3498 |
Other Identifiers: | ORCiD: Zhenchen Wang https://orcid.org/0000-0003-4710-0298 ORCiD: Allan Tucker https://orcid.org/0000-0001-5105-3506 |
Appears in Collections: | Dept of Computer Science Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
FullText.pdf | Copyright © 2021 B. Draghi, Z. Wang, P. Myles & A. Tucker. Published by ML Research Press. | 662.4 kB | Adobe PDF | View/Open |
Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.