Data type based aggregation of multiple linked tables for machine learning tasks

One or more non-transitory computer-readable storage media configured to store instructions that, in response to being executed, cause an electronic device to perform operations, the operations comprising:

• receiving relational data that includes a parent table and a set of linked tables associated with the parent table;
• executing a set of operations comprising:
• selecting a linked table from the set of linked tables;
• determining a data type that is associated with each column of the linked table;
• determining a set of transformations based on the data type;
• determining, from each column of the linked table, a set of features corresponding to each primary key in the parent table;
• generating one or more aggregate features corresponding to each primary key in the parent table by applying the set of transformations on the set of features;
• and
• updating an aggregate table to include the one or more aggregate features under one or more new columns of the aggregate table;
• repeating the execution of the set of operations on other linked tables of the set of tables to obtain the updated aggregate table;
• and
• training a machine learning model based on the updated aggregate table.

An electronic device, comprising:

• a memory configured to store instructions;
• and
• a processor, coupled to the memory, configured to execute the instructions to perform a process comprising:
• receiving relational data that includes a parent table and a set of linked tables associated with the parent table;
• executing a set of operations comprising:
• selecting a linked table from the set of linked tables;
• determining a data type that is associated with each column of the linked table;
• determining a set of transformations based on the data type;
• determining, from each column of the linked table, a set of features corresponding to each primary key in the parent table;
• generating one or more aggregate features corresponding to each primary key in the parent table by applying the set of transformations on the set of features;
• and
• updating an aggregate table to include the one or more aggregate features under one or more new columns of the aggregate table;
• repeating the execution of the set of operations on other linked tables of the set of tables to obtain the updated aggregate table;
• and
• training a machine learning model based on the updated aggregate table.