Is it possible to make statistical inference broadly accessible to non-statisticians without sacrificing mathematical rigor or inference quality? BayesDB is a probabilistic programming platform that enables users to query the probable implications of their data as directly as SQL databases enable them to query the data itself.
BayesDB also includes a domain general metamodel for machine assisted modeling of tabular structured data, controlled by MML, the Meta-Modeling Language. Machine assisted modeling is currently in private alpha.
We are actively seeing collaborators with interesting populations to model. Priority is given to individuals with data, domain expertise, and the time to work with us interactively on modeling the data and creating measures of inference quality.
Our preference is to work with people whose data can be published as useful examples, or which will be eventually published for peer review. However, we realize that many intersting data sets are private or contain sensitive information, and are open to those collaborations as well. Please include this information in your offer of collaboration.
BayesDB is currently in private alpha, only for collaborators. To suggest a collaboration, please fill out the BayesDB Alpha Request Form
Participants in the private alpha are expected to share data and work with our researchers to model the data, suggest compelling queries, use their domain expertise to validate model quality, and help us find bugs and usability issues.
Bug reports and feature should be submitted via Github bug tracking for Bayeslite.
If you have already been approved for the alpha and have your password, please proceed to the alpha site:
If you have any comments or questions, please feel free to email email@example.com.