Creating line item information from free-form tabular data

A system comprising:

• one or more computers;
• and
• a computer-readable medium coupled to the one or more computers having instructions stored thereon which, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
• receiving, at a system, event data values;
• identifying column headers of columns in the event data values, wherein at least one column header is not included in standard terms used by the system;
• identifying column values of the columns in the event data values;
• processing the identified column values using one or more content recognition models to generate a content match score for each pairing of a given column and a candidate standard term used by the system;
• processing the identified column headers using one or more context recognition models to generate a context match score for each pairing of a given column header and candidate standard term;
• creating an aggregate match score for each pairing of a column and candidate standard term based on the content match score for the pairing and the context match score for the pairing;
• determining a selected standard term for each column by identifying a candidate standard term paired with the column that has a highest aggregate match score;
• and
• creating an event in the system from the identified column values, wherein the event includes respective selected standard terms that were selected for respective columns.

A computer program product encoded on a non-transitory storage medium, the product comprising non-transitory, computer readable instructions for causing one or more processors to perform operations comprising:

• receiving, at a system, event data values;
• identifying column headers of columns in the event data values, wherein at least one column header is not included in standard terms used by the system;
• identifying column values of the columns in the event data values;
• processing the identified column values using one or more content recognition models to generate a content match score for each pairing of a given column and a candidate standard term used by the system;
• processing the identified column headers using one or more context recognition models to generate a context match score for each pairing of a given column header and candidate standard term;
• creating an aggregate match score for each pairing of a column and candidate standard term based on the content match score for the pairing and the context match score for the pairing;
• determining a selected standard term for each column by identifying a candidate standard term paired with the column that has a highest aggregate match score;
• and
• creating an event in the system from the identified column values, wherein the event includes respective selected standard terms that were selected for respective columns.