The Cantor Project

Cantor is a software project with the ambition to redefine transparency in data processing; from business reporting to preparation of data for statistical analysis and machine learning.

As more and more processes are being automated, transparency in data preparation is heavily reduced. Organizations increasingly rely on information and automated recommendations that only a few can explain in detail. The issue of transparency is acute and highlighted by demands for new legislations and even lawsuits due to non-transparent data processing*.

The programming language and associated web services developed within the Cantor project will enable multidisciplinary teams to understand even extensive data preparation processes. Cantor reduces the risk of "unverified" information and increases the ability to co-create complex processes.

Cantor will also make it easier to share data models between developers and teams; which fundamentally increases an organisation’s rate of learning and development.

* More on the issues of non-transparency within data processing and automated decisions

Stoyanovich & Howe: Follow the Data! Algorithmic Transparency Starts with Data Transparency
https://ai.shorensteincenter.org/ideas/2018/11/26/follow-the-data-algorithmic-transparency-starts-with-data-transparency

James Vacca et al: A Local Law in relation to automated decision systems used by agencies (File # Int 1696-2017, Law number: 2018/049)
https://legistar.council.nyc.gov/LegislationDetail.aspx?ID=3137815&GUID=437A6A6D-62E1-47E2-9C42-461253F9C6D0

TT: Robot decision in the social service reported to the Ombudsman
https://www.svd.se/robotbeslut-i-socialtjansten-jo-anmals (NB: article in Swedish)

The Cantor Project

Creating transparency in data preparation and data modelling

Background

Project members

Cantor lab cases

* More on the issues of non-transparency within data processing and automated decisions