Extracting purchase-related information from digital documents

A system, comprising a non-transitory memory storing processor-readable instructions, and a processor coupled to the memory, operable to execute the instructions, and based at least in part on the execution of the instructions operable to perform operations comprising:

• identifying electronic messages concerning respective product orders in respective collections of electronic messages transmitted between network nodes and stored in one or more non-transitory networked electronic message storage systems;
• for each of respective ones of the identified electronic messages,
• selecting a respective extraction parser corresponding to the identified electronic message, wherein each extraction parser is generated from a respective template defining a respective context-free grammar that describes a respective format of an electronic message body comprising an ordered arrangement of unstructured free form text elements, and
• applying, by a processor, the selected extraction parser to unstructured free form text in a body portion of the identified electronic message to extract therefrom a respective set of product order related data values corresponding to a predefined set of purchase order related data field types;
• storing the sets of extracted product order related data values as aggregated data in a non-transitory networked data storage system;
• for each of respective ones of the sets of extracted product order related data values, retrieving additional information concerning product order information missing from the respective set of extracted product order related data values based on one or more of the extracted product order related data values in the respective set of extracted product order related data values;
• and
• augmenting the aggregated data with the retrieved additional information.

At least one non-transitory computer readable medium comprising instructions, which when executed by computer apparatus, cause the computer apparatus to perform operations comprising:

• identifying electronic messages concerning respective product orders in respective collections of electronic messages transmitted between network nodes and stored in one or more non-transitory networked electronic message storage systems;
• for each of respective ones of the identified electronic messages,
• selecting a respective extraction parser corresponding to the identified electronic message, wherein each extraction parser is generated from a respective template defining a respective context-free grammar that describes a respective format of an electronic message body comprising an ordered arrangement of unstructured free form text elements, and
• applying, by a processor, the selected extraction parser to unstructured free form text in a body portion of the identified electronic message to extract therefrom a respective set of product order related data values corresponding to a predefined set of purchase order related data field types;
• storing the sets of extracted product order related data values as aggregated data in a non-transitory networked data storage system;
• for each of respective ones of the sets of extracted product order related data values, retrieving additional information concerning product order information missing from the respective set of extracted product order related data values based on one or more of the extracted product order related data values in the respective set of extracted product order related data values;
• and
• augmenting the aggregated data with the retrieved additional information.