The automated data analyst (ADAN) is a project, in Java, for automatically cleaning up and analysing data.
The following have been implemented:
-
DataFrame: A DataFrame implementation for Java, based on R's dataframe and Pandas DataFrame for python
-
R interface: ADAN can interface with R in order to run analyses
-
Drools: ADAN implements Drools as the underlying expert system for deciding how to clean and analyze the data.
ADAN is not fully working yet. However, the architecture and important test cases have been set in place. The system is designed in a way that it is possible to interface with other languages or libraries besides R. Adding new AI functionality for cleaning up or analysing data is as easy as simply writing up a new drools file.
For more, please read the documentation attached. If you have any questions please contact me directly.