FORM VALUE PREDICTION UTILIZING SYNONYMOUS FIELD RECOGNITION

• US 20170046622A1
• Filed: 08/12/2015
• Published: 02/16/2017
• Est. Priority Date: 08/12/2015
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: GAITHER, SHAWN ALAN, STROH, EYLON, MATHUR, PRIYANK, SWINEFORD, RANDY
• Granted Patent: US 10,133,813 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms

Abstract

Embodiments of the present invention provide systems, methods, and computer storage media directed at predicting values for an electronic form. In embodiments, the method can include forming synonym groupings of form field labels for a number of users. The synonym groupings can be based on an analysis of the similarity of form field values that are associated with form field labels. In embodiments a predictive model may be generated from these synonym groupings. The predictive model can correlate the synonym groupings of one user with synonym groupings of one or more additional users to enable a determination of one or more predicted form field values for the one user based on a queried form field label even though the one user may have never submitted an electronic form with the queried form field label. Other embodiments may be described and/or claimed.

Claims

1. One or more computer-readable media having instructions embodied thereon, wherein the instructions, when executed by one or more processors of a computing device, enable the computing device to predict values for electronic forms by causing the one or more processors to:

receive a form field label associated with an electronic form being completed by a user of the computing device, wherein the form field label is absent from a form field profile associated with the user;

correlate the received form field label with one or more synonymous form field labels; and

determine at least one predicted form field value for the user, based on the one or more synonymous form field labels and the form field profile associated with the user.

2. The one or more computer-readable media of claim 1, wherein to correlate the received form field label with one or more synonymous form field labels is accomplished utilizing a predictive model that correlates first synonym groupings of the user with one or more additional synonym groupings of one or more additional users.

3. The one or more computer-readable media of claim 1, wherein to determine at least one predicted form field value of the received form field label is to determine a plurality of predicted form field values and cause the plurality of predicted form field values to be presented to the user for selection of one of the predicted form field values.

4. The one or more computer-readable media of claim 3, wherein an order of the plurality of predicted form field values is based on a confidence level associated with each of the plurality of predicted form field values.

5. The one or more computer-readable media of claim 3, wherein an order of the plurality of predicted form field values is based on one or more other form field values of a form being completed by the user.

6. The one or more computer-readable media of claim 1, wherein an order of the plurality of predicted form field values is based on a confidence level associated with each of the plurality of predicted form field values, wherein the confidence level is based on a measure of the number of synonymous form field labels that are associated with each of the predicted form field values.

7. A computer-implemented method of predicting values for an electronic form comprising:

forming first synonym groups of first form field labels for a first user based on an analysis of similarity of form field values that are associated with the first form field labels, and that have been submitted by the first user; and

generating a predictive model to correlate the first synonym groupings with one or more additional synonym groupings of one or more additional users, the correlating based, at least in part, on the first form field labels and additional form field labels contained within the one or more additional synonym groupings, wherein the predictive model enables determination of one or more predicted form field values for the first user based on a queried form field label that exists in the one or more additional form field labels but is absent from the first form field labels.

8. The method of claim 7, wherein forming first synonym groupings of the first form field labels based on an analysis of similarity of the form field values comprises:

generating a first graph including label nodes for each of the first form field labels, value nodes for each of the form field values, and edges connecting the label nodes with respective value nodes; and

combining value nodes of the first graph that are within a level of similarity from one another into a single node.

9. The method of claim 8, wherein correlating the first groupings with the one or more additional groupings comprises incorporating the first groupings and the one or more additional groupings into a global graph.

10. The method of claim 7, wherein generating the predictive model comprises performing a PageRank process to determine a similarity between the first form field labels and the additional form field labels.

11. The method of claim 7, wherein forming the first synonym groupings of the first form field labels further comprises normalizing the first form field labels.

12. The method of claim 7, wherein forming the first synonym groupings of the first form field labels further comprises deduplicating the first form field labels.

13. The method of claim 7, further comprising:

obfuscating the form field values in a manner specific to the first user.

14. The method of claim 13, wherein obfuscating the form field values includes encryption of the form field values utilizing an encryption key specific to the first user.

15. The method of claim 13, wherein obfuscating the form field values includes hashing of the form field values utilizing a seed specific to the first user.

16. The method of claim 7, further comprising:

obfuscating the first form field labels in a user independent manner, wherein the additional form field labels are obfuscated in the same manner such that the obfuscated first form field labels and the obfuscated additional form field labels are capable of comparison against one another based on similarity.

17. A computing system for predicting values for electronic forms comprising:

one or more processors;

a field training engine that is to cause the one or more processors to;

generate a predictive model that correlates a plurality of form field labels based on a plurality of synonym groupings that are each associated with a respective user of a plurality of users;

a field recognition engine that is to cause the one or more processors to;

utilize the predictive model to determine one or more form field labels that are synonymous with a queried form field label of an electronic form being completed by one user, of the plurality of users, wherein the queried form field label is absent from a form field profile of the one user; and

determine at least one predicted form field value associated with the one user for the queried form field label utilizing the synonymous one or more form field labels and a form field profile of the one user.

18. The computing system of claim 17, wherein each of the plurality of synonym groupings includes a group of form field labels of the plurality of form field labels that have been correlated based on an analysis of similarity between form field values submitted by the respectively associated user in response to each of the form field labels of the group of form field labels.

19. The computing system of claim 17, wherein to determine at least one predicted form field value is to determine a plurality of predicted form field values and cause the plurality of predicted form field values to be presented to the user for selection of one of the predicted form field values.

20. The computing system of claim 19, wherein an order of the plurality of predicted form field values is based on a confidence level associated with each of the plurality of predicted form field values.