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

Information retrieval (IR) systems often leverage query data to suggestrelevant items to users. This introduces the possibility of unfairness if thequery (i.e., input) and the resulting recommendations unintentionally correlatewith latent factors that are protected variables (e.g., race, gender, and age).For instance, a visual search system for fashion recommendations may pick up onfeatures of the human models rather than fashion garments when generatingrecommendations. In this work, we introduce a statistical test for"distribution parity" in the top-K IR results, which assesses whether a givenset of recommendations is fair with respect to a specific protected variable.We evaluate our test using both simulated and empirical results. First, usingartificially biased recommendations, we demonstrate the trade-off betweenstatistically detectable bias and the size of the search catalog. Second, weapply our test to a visual search system for fashion garments, specificallytesting for recommendation bias based on the skin tone of fashion models. Ourdistribution parity test can help ensure that IR systems' results are fair andproduce a good experience for all users.

Recommendation or Discrimination?: Quantifying Distribution Parity in Information Retrieval Systems