English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning models built on datasets containing discriminative instancesattributed to various underlying factors result in biased and unfair outcomes.It's a well founded and intuitive fact that existing bias mitigation strategiesoften sacrifice accuracy in order to ensure fairness. But when AI engine'sprediction is used for decision making which reflects on revenue or operationalefficiency such as credit risk modelling, it would be desirable by the businessif accuracy can be somehow reasonably preserved. This conflicting requirementof maintaining accuracy and fairness in AI motivates our research. In thispaper, we propose a fresh approach for simultaneous improvement of fairness andaccuracy of ML models within a realistic paradigm. The essence of our work is adata preprocessing technique that can detect instances ascribing a specifickind of bias that should be removed from the dataset before training and wefurther show that such instance removal will have no adverse impact on modelaccuracy. In particular, we claim that in the problem settings where instancesexist with similar feature but different labels caused by variation inprotected attributes , an inherent bias gets induced in the dataset, which canbe identified and mitigated through our novel scheme. Our experimentalevaluation on two open-source datasets demonstrates how the proposed method canmitigate bias along with improving rather than degrading accuracy, whileoffering certain set of control for end user.

Simultaneous Improvement of ML Model Fairness and Performance by Identifying Bias in Data