Data-driven algorithms are being studied and deployed in diverse domains to support decisions. They directly impact on people's life and for this reason more and more researchers are analyzing the equity of existing algorithms and they are proposing new ones. Algorithmic fairness progress is based on data, which can be used in a correct way only if sufficiently documented. As stated by Fabris et al. (2022) "Unfortunately, the algorithmic fairness community, suffers from a collective data documentation debt caused by a lack of information on specific resources and scatteredness of available information.". This thesis is strictly connected with the researches done in this field by Prof. Gianmaria Silvello and it proposes a Web Application to support and share their work.
Gli algoritmi basati sui dati vengono studiati e utilizzati in diversi ambiti per supportare le decisioni. Questi impattano direttamente sulla vita delle persone e per questo motivo sempre più ricercatori analizzano l'equità degli algoritmi esistenti e ne propongono di nuovi. L'equità algoritmica si basa sui dati, che possono essere utilizzati in modo corretto solo se sufficientemente documentati. Come riportato da Fabris et al. (2022) "Purtroppo, in questo ambito si soffre di un debito collettivo di documentazione dei dati, causato dalla mancanza di informazioni su risorse specifiche e dalla dispersione delle informazioni disponibili.". Questa tesi è strettamente legata alle ricerche condotte in questo campo dal Prof. Gianmaria Silvello et al. e propone una Web Application per supportare e condividere il loro lavoro.
A Web Application for Searching Fairness Datasets
PIVA, ALBERTO
2021/2022
Abstract
Data-driven algorithms are being studied and deployed in diverse domains to support decisions. They directly impact on people's life and for this reason more and more researchers are analyzing the equity of existing algorithms and they are proposing new ones. Algorithmic fairness progress is based on data, which can be used in a correct way only if sufficiently documented. As stated by Fabris et al. (2022) "Unfortunately, the algorithmic fairness community, suffers from a collective data documentation debt caused by a lack of information on specific resources and scatteredness of available information.". This thesis is strictly connected with the researches done in this field by Prof. Gianmaria Silvello and it proposes a Web Application to support and share their work.File | Dimensione | Formato | |
---|---|---|---|
Piva_Alberto.pdf
accesso aperto
Dimensione
25.54 MB
Formato
Adobe PDF
|
25.54 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/40289