Determining user personality characteristics from social networking system communications and characteristics

• US 9,740,752 B2
• Filed: 06/03/2016
• Issued: 08/22/2017
• Est. Priority Date: 09/10/2012
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Nowak, Michael, Eckles, Dean
• Primary Examiner(s): KATSIKIS, KOSTAS J
• Application Number:
  US15/173,009
• Publication Number:
  US 20160283485A1
• Time in Patent Office: 445 Days
• Field of Search: 709204
• CPC Class Codes: G06F 15/16 Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs coordinating program control therefor G06F9/52; in regulating and control system G05B
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/52 for supporting social networking services
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/306 User profiles

Abstract

A social networking system obtains linguistic data from a user'"'"'s text communications on the social networking system. For example, occurrences of words in various types of communications by the user in the social networking system are determined. The linguistic data and non-linguistic data associated with the user are used in a trained model to predict one or more personality characteristics for the user. The inferred personality characteristics are stored in connection with the user'"'"'s profile, and may be used for targeting, ranking, selecting versions of products, and various other purposes.

Claims

1. A computer-implemented method comprising:

extracting, by a communication network, linguistic data from at least one type of communication between a user of the communication network and one or more additional users of the communication network;

retrieving at least one characteristic of the user from a user profile of the user at the communication network;

applying at least one statistical model to the extracted linguistic data and the at least one retrieved characteristics of the user, the at least one statistical model being determined by;

determining one or more personality characteristics of a training set of users, the one or more personality characteristics being determined based on responses to one or more surveys received from the training set of users; and

generating the at least one statistical model based on the determined one or more personality characteristics and linguistic data retrieved from user profiles associated with the training set of users at the communication network;

selecting at least one personality characteristics for the user, the selected at least one personality characteristic being associated with at least a threshold value from the at least one statistical model;

storing the at least one selected personality characteristic in the user profile of the user; and

presenting content to the user based at least in part on the at least one selected personality characteristic.

2. The computer-implemented method of claim 1, wherein the at least one statistical model is further generated by:

providing the one or more surveys to the training set of users of the communication network, the one or more surveys including a plurality of items associated with the one or more personality characteristics; and

receiving responses to the one or more surveys from responding users in the training set.

3. The computer-implemented method of claim 2, wherein presenting the content to the user based at least in part on the determined one or more personality characteristics of the user comprises:

selecting one or more stories for inclusion in a news feed presented to the user based at least in part on the determined one or more personality characteristics of the user; and

presenting the selected one or more stories to the user in the news feed.

4. The computer-implemented method of claim 2, wherein presenting content to the user based at least in part on the determined one or more personality characteristics comprises:

selecting one or more advertisements for presentation to the user based at least in part on the determined one or more personality characteristics; and

presenting the selected one or more advertisements to the user.

5. The computer-implemented method of claim 4, wherein selecting one or more advertisements for presentation to the user based at least in part on the determined one or more personality characteristics comprises:

selecting advertisements associated with one or more targeting criteria matching at least one of the determined personality characteristics.

6. The computer-implemented method of claim 1, wherein one or more of the statistical models perform at least a rank correlation analysis of the extracted linguistic data correlating personality characteristics and word stem category proportions from the linguistic data.

7. The computer-implemented method of claim 1, wherein the at least one type of communication from which the linguistic data is extracted includes at least one of status updates, notes, messages, posts, or comments.

8. The computer-implemented method of claim 7, wherein extracting the linguistic data from the at least one type of communication comprises:

determining a count of words associated with a plurality of categories in a data set, wherein a count of word stems in each of the categories is used as the linguistic data for the user.

9. The computer-implemented method of claim 1, wherein presenting content to the user based at least in part on the at least one selected personality characteristic comprises:

selecting one or more recommendations for actions to the user based at least in part on the at least one selected personality characteristic; and

presenting the selected one or more recommendations for actions to the user.

10. The computer-method of claim 1, wherein the at least one characteristic of the user includes at least one of an age, a gender, a number of additional users connected to the user, a percentage of connections to other users initiated by the user, a presence of a profile picture in the user profile, a number of times the user accesses the social networking system within a specified time interval, a number of communications from the user having different communication types, a frequency with which the user creates different communication types, a total number of communications generated by the user, a percentage of communications generated by the user having different types of communication, or a number of unique days that the user has generated communications having various types of communication.

11. A non-transitory computer-readable storage medium including instructions that, when executed by a processor, cause the processor to:

extract linguistic data from at least one type of communication between a user of a communication network and one or more additional users of the communication network;

retrieve at least one characteristic of the user from a user profile of the user at the communication network

apply at least one statistical model to the extracted linguistic data and the at least one retrieved characteristics of the user, the at least one statistical model being determined by;

determining one or more personality characteristics of a training set of users, the one or more personality characteristics being determined based on responses to one or more surveys received from the training set of users; and

generating the at least one statistical model based on the determined one or more personality characteristics and linguistic data retrieved from user profiles associated with the training set of users at the communication network;

select at least one personality characteristics for the user, the selected at least one personality characteristic being associated with at least a threshold value from the at least one statistical model;

store the at least one selected personality characteristic in the user profile of the user; and

present content to the user based at least in part on the at least one selected personality characteristic.

12. The non-transitory computer readable storage medium of claim 11, wherein the at least statistical model is further generated by:

providing the one or more surveys to the training set of users of the communication network, the one or more surveys including a plurality of items associated with the one or more personality characteristics; and

receiving responses to the one or more surveys from responding users in the training set.

13. The non-transitory computer readable storage medium of claim 11, wherein one or more of the statistical models perform at least a rank correlation analysis of the extracted linguistic data correlating personality characteristics and word stem category proportions from the linguistic data.

14. The non-transitory computer readable storage medium of claim 11, wherein the at least one type of communication from which the linguistic data is extracted includes at least one of status updates, notes, messages, posts, or comments.

15. The non-transitory computer readable storage medium of claim 11, wherein extracting the linguistic data from the at least one type of communication comprises:

determining a count of words associated with a plurality of categories in a data set, wherein a count of word stems in each of the categories is used as the linguistic data for the user.