Athenic AI's query generation performance is affected by how “clean” and organized a user’s database is. In some cases, column names are heavily abbreviated making it difficult to extract semantic meaning from the name. Other times, tables have many columns that are not relevant to your particular data analyses, which just becomes noise for both you and our AI. Athenic AI’s Datasets feature allows users with more complex databases to improve the AI SQL generation without having to make changes directly to their own live, production databases.
Datasets is an advanced feature of Athenic AI for users who have familiarity with SQL. Datasets are separate read-only views of your database and does not store any actual data from your database. It requires users to write SQL in order to generate the “cleaned” views of your database (e.g. rename columns or combine relevant tables into one). Only this SQL query will be saved by us.
For Enterprise users, we offer datasets services to create your datasets to provide a better experience using Athenic AI! To upgrade to Enterprise or for more information, please do not hesitate to schedule a meeting.
To learn more about datasets, check out our documentation.