Systems and methods for generating group node profiles based on member nodes

Patent metadata

Field	Value
Publication number	US-10878015-B2
Application number	US-201916398150-A
Country	US
Kind code	B2
Filing date	Apr 29, 2019
Priority date	May 24, 2018
Publication date	Dec 29, 2020
Grant date	Dec 29, 2020

Abstract

The present disclosure relates to a method for generating a group node profile based on member nodes. Member node profiles may be maintained. A subset of member node profiles may be identified. Each member node profile of the subset may correspond to a group entity. A set of member node fields to populate values for a group node profile of the group entity may be identified. A corresponding group node field of the group node profile may be updated in accordance with a group node population policy. A field value of the member node field to the group node field may be added. An entry to a value data structure of the field value of the group node field may be added.

First claim

What is claimed is:

1. A method, comprising:
maintaining, by one or more processors, a plurality of member node profiles corresponding to a plurality of member entities, each member node profile having a plurality of member node fields, each member node field having one or more field values extracted from at least one of an electronic activity or at least one record object of a plurality of systems of record;
identifying, by the one or more processors, from the plurality of member node profiles, a subset of member node profiles, each member node profile of the subset corresponding to a group entity;
identifying, by the one or more processors, from the plurality of member node fields of each member node profile of the subset, a set of member node fields to populate values for corresponding group node fields of a group node profile of the group entity;
maintaining, by the one or more processors, for each field value of each member node field, a confidence score indicating a likelihood that the field value is accurate;
for each member node field of the set of member node fields,
i) determining, by the one or more processors, that the respective confidence score of a corresponding field value of the member node field of the member node profile satisfies a threshold score by comparing the confidence score of the field value of the member node field to the threshold score; and
ii) updating, by the one or more processors, in accordance with a group node population policy and responsive to the confidence score of the field value of the member node field satisfying the threshold score, a group node field of the group node profile corresponding to the member node field by adding at least one of the field value of the member node field as a new value to the group node field or adding an entry to a value data structure of the field value of the group node field; and

determining, by the one or more processors, for each field value of each group node field of the group node profile, a respective confidence score for the field value based on one or more confidence scores of the corresponding field values of the set of member node fields.

2. The method of claim 1 , wherein each member node profile of the subset corresponding to the group entity includes at least one field-value pair matching a field-value pair of other member node profiles in the subset.

3. The method of claim 1 , wherein each member node profile of the subset corresponding to the group entity includes at least one field-value pair matching a field-value pair of a predetermined portion of field-value pairs of other member node profiles in the subset.

4. The method of claim 1 , wherein identifying the set of member node fields further comprises identifying the set of member node fields corresponding to predetermined member node fields in accordance with the group node population policy, the group node population policy specifying the predetermined member node fields.

5. The method of claim 1 , wherein the threshold score is predetermined in accordance to the group node population policy, the group node population policy specifying a plurality of threshold scores for the plurality of corresponding member node fields.

6. The method of claim 1 , wherein the confidence score of the field value of each member node field is generated based on a number of data points included in the field value of each member node field.

7. A method, comprising:
maintaining, by one or more processors, a plurality of member node profiles corresponding to a plurality of member entities, each member node profile having a plurality of member node fields, each member node field having one or more field values extracted from at least one of an electronic activity or at least one record object of a plurality of systems of record;
identifying, by the one or more processors, from the plurality of member node profiles, a subset of first member node profiles, each first member node profile of the subset corresponding to a group entity;
identifying, by the one or more processors, from the plurality of member node fields of each first member node profile of the subset, a set of member node fields to populate field values for corresponding group node fields of a group node profile of the group entity;
for each member node field of the set of member node fields of each first member node profile; updating, by the one or more processors, in accordance with a group node population policy, a corresponding group node field of the group node profile by adding at least one of a field value of the member node field to the group node field or an entry to a value data structure of the field value of the group node field;
determining, by the one or more processors, for the field value of the group node field of the group node profile, a respective confidence score for the field value based on one or more confidence scores of the corresponding field values of the set of member node fields;
identifying by the one or more processors, a second member node profile corresponding to the group node profile;
determining, by the one or more processors, to populate, a member node field of the second member node profile by determining that the member node field of the second node profile is either i) missing a field value, or ii) includes a field value different from the field value of the corresponding group node field in the group node profile; and
updating, by the one or more processors, the member node field of the second member node profile to include as a value, the field value of the corresponding group node field in the group node profile.

8. A method, comprising:
maintaining, by one or more processors, a plurality of member node profiles corresponding to a plurality of member entities, each member node profile having a plurality of member node fields, each member node field having one or more field values extracted from at least one of an electronic activity or at least one record object of a plurality of systems of record;
identifying, by the one or more processors, from the plurality of member node profiles, a subset of member node profiles, each member node profile of the subset corresponding to a group entity;
identifying, by the one or more processors, from the plurality of member node fields of each member node profile of the subset, a set of member node fields to populate values for a group node profile of the group entity;
for each member node field of the set of member node fields of each member node profile, updating, by the one or more processors, in accordance with a group node population policy, a corresponding group node field of the group node profile by adding at least one of a field value of the member node field to the group node field or an entry to a value data structure of the field value of the group node field;
identifying, by the one or more processors, a particular electronic activity that is the electronic activity or a second electronic activity transmitted or received from an electronic account identifier of a particular member node profile of the plurality of member node profiles;
determining, by the one or more processors, that the particular electronic activity is associated with at least one member node profile of the subset of member node profiles corresponding to the group entity;
extracting, by the one or more processors, responsive to parsing the particular electronic activity, a second field value from the particular electronic activity;
matching, by the one or more processors, the particular electronic activity to a member node profile of the subset of member node profiles;
updating, by the one or more processors, the group node profile by adding the second field value as a value of the corresponding group

Assignees

• People Ai Inc

Inventors

• Rogynskyy Oleg
• Brunets Yurii
• Hermanek Stefan
• Markovsky Yury

Classifications

• G06N3/08

  Learning methods · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06F16/2358

  Change logging, detection, and notification (replication G06F16/27) · CPC title

• G06F9/542

  Event management; Broadcasting; Multicasting; Notifications · CPC title

• G06Q10/1093

  Calendar-based scheduling for persons or groups · CPC title