System and method for selecting data sample groups for machine learning of context of data fields for various document types and/or for test data generation for quality assurance systems

A non-transitory computer-readable medium having a plurality of computer-executable instructions which, when executed by a processor, perform a method for efficiently learning new forms in an electronic document preparation system, the instructions comprising:

• an interface module configured to receive form data representing to a new form having a plurality of data fields;
• a data acquisition module configured to gather training set data related to previously filled forms, each previously filled form having completed data fields that each correspond to a respective data field of the new form;
• a grouping module configured to generate, for each selected data field of the new form, grouping data by forming a plurality of groups from the training set data based on respective categories and assigning each of a plurality of the previously filled forms to a respective one of the groups;
• a sampling module configured to generate, for each selected data field of the new form, sampling data by selecting one or more previously filled forms from each group of the grouping data associated with the selected data field;
• a machine learning module configured to generate, for each selected data field, candidate function data relating to a plurality of candidate functions, to generate, for each selected data field, test data by applying the candidate functions to the training set data in accordance with the sampling data, and to identify, for each selected data field, an acceptable function from the plurality of candidate functions based on a how closely the test data matches the candidate function data.

A system for efficiently learning new forms in an electronic document preparation system, the system comprising:

• at least one processor;
• and
• at least one memory coupled to the at least one processor, the at least one memory having stored therein instructions which, when executed by any set of the one or more processors, perform a process including:
• receiving, with an interface module of a computing system, form data related to a new form having a plurality of data fields;
• gathering, with a data acquisition module of a computing system, training set data related to previously filled forms, each previously filled form having completed data fields that each correspond to a respective data field of the new form;
• generating, with a grouping module of a computing system and for a first selected data field of the new form, grouping data by forming a plurality of groups from the training set data based on respective categories and assigning each of a plurality of the previously filled forms to a respective one of the groups;
• generating, with a sampling module of a computing system, sampling data by selecting one or more previously filled forms from each group;
• generating, with a machine learning module of a computing system, for the first selected data field, candidate function data including a plurality of candidate functions;
• generating, with the machine learning module and for each candidate function, test data by applying the candidate function to a portion of the training set data corresponding to the sampling data;
• identifying, with the machine learning module and from the plurality of functions, an acceptable candidate for the first selected data field, by comparing the test data to the training set data and identifying test data that matches the training set data within a selected tolerance;
• generating, with the machine learning module, results data indicating the acceptable function for the first data field of the new form;
• and
• outputting, with the interface module, the results data.

A computing system implemented method for grouping and sampling data sets, the method comprising:

• gathering training set data related to previously filled forms each having a plurality of data fields;
• generating, for a first selected data field of the plurality of data fields, grouping data by forming a plurality of groups from the training set data based on respective categories and assigning each of a plurality of the previously filled forms to a respective one of the groups based on the categories;
• generating, for the first selected data field, sampling data by selecting one or more previously filled forms from each group, wherein the groups are selected to ensure that the sampling data will include previously filled forms having uncommon data values in data fields corresponding to the selected data field or in data fields included in one or more of the candidate functions;
• and
• providing a portion of the training set data corresponding to the sampling data to a quality assurance system.