Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386242314> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4386242314 abstract "Past research has demonstrated that the explicit use of protected attributes in machine learning can improve both performance and fairness. Many machine learning algorithms, however, cannot directly process categorical attributes, such as country of birth or ethnicity. Because protected attributes frequently are categorical, they must be encoded as features that can be input to a chosen machine learning algorithm, e.g. support vector machines, gradient boosting decision trees or linear models. Thereby, encoding methods influence how and what the machine learning algorithm will learn, affecting model performance and fairness. This work compares the accuracy and fairness implications of the two most well-known encoding methods: one-hot encoding and target encoding. We distinguish between two types of induced bias that may arise from these encoding methods and may lead to unfair models. The first type, irreducible bias, is due to direct group category discrimination and the second type, reducible bias, is due to the large variance in statistically underrepresented groups. We investigate the interaction between categorical encodings and target encoding regularization methods that reduce unfairness. Furthermore, we consider the problem of intersectional unfairness that may arise when machine learning best practices improve performance measures by encoding several categorical attributes into a high-cardinality feature." @default.
- W4386242314 created "2023-08-30" @default.
- W4386242314 creator A5058755772 @default.
- W4386242314 creator A5062807811 @default.
- W4386242314 creator A5071524745 @default.
- W4386242314 creator A5081978740 @default.
- W4386242314 date "2023-08-08" @default.
- W4386242314 modified "2023-10-06" @default.
- W4386242314 title "Fairness Implications of Encoding Protected Categorical Attributes" @default.
- W4386242314 cites W1961345416 @default.
- W4386242314 cites W1990836268 @default.
- W4386242314 cites W1999380087 @default.
- W4386242314 cites W2014352947 @default.
- W4386242314 cites W2026019770 @default.
- W4386242314 cites W2100960835 @default.
- W4386242314 cites W2106100548 @default.
- W4386242314 cites W2116984840 @default.
- W4386242314 cites W2165960883 @default.
- W4386242314 cites W2166454173 @default.
- W4386242314 cites W2282821441 @default.
- W4386242314 cites W2392403817 @default.
- W4386242314 cites W2584805976 @default.
- W4386242314 cites W2799900537 @default.
- W4386242314 cites W2810511505 @default.
- W4386242314 cites W2901823434 @default.
- W4386242314 cites W2954052001 @default.
- W4386242314 cites W2963104135 @default.
- W4386242314 cites W2963105378 @default.
- W4386242314 cites W2997591727 @default.
- W4386242314 cites W3020873385 @default.
- W4386242314 cites W3023702633 @default.
- W4386242314 cites W3101427066 @default.
- W4386242314 cites W3102476541 @default.
- W4386242314 cites W3118480630 @default.
- W4386242314 cites W3135468018 @default.
- W4386242314 cites W3142848995 @default.
- W4386242314 cites W3202428668 @default.
- W4386242314 cites W4233413206 @default.
- W4386242314 cites W4283164653 @default.
- W4386242314 cites W4283166422 @default.
- W4386242314 cites W4289258088 @default.
- W4386242314 doi "https://doi.org/10.1145/3600211.3604657" @default.
- W4386242314 hasPublicationYear "2023" @default.
- W4386242314 type Work @default.
- W4386242314 citedByCount "0" @default.
- W4386242314 crossrefType "proceedings-article" @default.
- W4386242314 hasAuthorship W4386242314A5058755772 @default.
- W4386242314 hasAuthorship W4386242314A5062807811 @default.
- W4386242314 hasAuthorship W4386242314A5071524745 @default.
- W4386242314 hasAuthorship W4386242314A5081978740 @default.
- W4386242314 hasBestOaLocation W43862423141 @default.
- W4386242314 hasConcept C119857082 @default.
- W4386242314 hasConcept C12267149 @default.
- W4386242314 hasConcept C124101348 @default.
- W4386242314 hasConcept C125411270 @default.
- W4386242314 hasConcept C154945302 @default.
- W4386242314 hasConcept C41008148 @default.
- W4386242314 hasConcept C46686674 @default.
- W4386242314 hasConcept C5274069 @default.
- W4386242314 hasConcept C87117476 @default.
- W4386242314 hasConceptScore W4386242314C119857082 @default.
- W4386242314 hasConceptScore W4386242314C12267149 @default.
- W4386242314 hasConceptScore W4386242314C124101348 @default.
- W4386242314 hasConceptScore W4386242314C125411270 @default.
- W4386242314 hasConceptScore W4386242314C154945302 @default.
- W4386242314 hasConceptScore W4386242314C41008148 @default.
- W4386242314 hasConceptScore W4386242314C46686674 @default.
- W4386242314 hasConceptScore W4386242314C5274069 @default.
- W4386242314 hasConceptScore W4386242314C87117476 @default.
- W4386242314 hasLocation W43862423141 @default.
- W4386242314 hasOpenAccess W4386242314 @default.
- W4386242314 hasPrimaryLocation W43862423141 @default.
- W4386242314 hasRelatedWork W1540371141 @default.
- W4386242314 hasRelatedWork W2125652721 @default.
- W4386242314 hasRelatedWork W2154063878 @default.
- W4386242314 hasRelatedWork W2333896121 @default.
- W4386242314 hasRelatedWork W2400708317 @default.
- W4386242314 hasRelatedWork W2556012038 @default.
- W4386242314 hasRelatedWork W3114793362 @default.
- W4386242314 hasRelatedWork W4229333355 @default.
- W4386242314 hasRelatedWork W4231274751 @default.
- W4386242314 hasRelatedWork W4312820300 @default.
- W4386242314 isParatext "false" @default.
- W4386242314 isRetracted "false" @default.
- W4386242314 workType "article" @default.