System, method, and computer program for identifying message content to send to users based on user language characteristics

1. A method, performed by a computer system, for identifying message content to send to users based on the users' language characteristics, the method comprising:

obtaining user-generated content for a plurality of users, wherein each content item is associated with a user ID that uniquely identifies the user that generated the content item;

extracting language characteristics from the user-generated content;

assigning one or more language characteristic scores to each of the users;

clustering the users into groups using the language characteristic scores;

testing the responsiveness of each group to different message content by sending a plurality of test messages with different message content to at least a subset of users within each group and measuring the response rates to the test messages;

associating different message content with different language characteristic scores by identifying a message content to which each of the groups is most responsive and associating the message content with the group;

obtaining user-generated content for a new user;

extracting language characteristics from the new user's user-generated content;

assigning one or more language characteristic scores to the new user; and

sending a message to the new user with the message content previously associated with users having similar language characteristic score(s).

2. The method of claim 1 , wherein testing the responsiveness of each group to different message content comprises testing the responsiveness of each group to offers of different products to identify a best product for each group.

3. The method of claim 2 , wherein, after a best product for each group is identified, testing the responsiveness of each group to certain vocabularies by sending test messages to at least a subset of users in each group that include an offer for the best product for the group and the certain vocabularies.

4. The method of claim 3 , wherein, for each group, the message content associated with the group comprises the best product for the group and the vocabulary to which the group is most responsive.

5. The method of claim 1 , wherein associating a message content with a group comprises associating a certain vocabulary with the group.

6. The method of claim 1 , wherein associating a message content with the group comprises associating certain vocabulary rules with the group.

7. The method of claim 1 , wherein the steps pertaining to the new user are performed on a client device of the new user by a client application that executes rules that map each group to a select message.

8. The method of claim 1 , wherein the language characteristics on which the users are scored include type of vocabulary used.

9. The method of claim 8 , wherein the language characteristics on which the users are scored also include morphology characteristics of the language.

10. The method of claim 9 , wherein the morphology characteristics are one of: punctuation, capitalization, spelling errors, grammar errors, and acronyms.

11. The method of claim 1 , wherein the language characteristics on which the users are scored include use of emoticons.

12. A system for identifying message content to send to users based on the users' language characteristics, the system comprising:

a server configured to perform the following:

obtaining user-generated content for a plurality of users, wherein each content item is associated with a user ID that uniquely identifies the user that generated the content item;

extracting language characteristics from the user-generated content;

assigning one or more language characteristic scores to each of the users;

clustering the users into groups using the language characteristic scores;

creating group definitions for each of the groups;

testing the responsiveness of each group to different message content by sending a plurality of test messages with different message content to at least a subset of users within each group and measuring the response rates to the test messages;

associating different message content with different language characteristic scores by identifying a message content to which each of the groups is most responsive and creating rules for each group based on the message content; and

a client device configured to perform the following:

obtaining the group definitions from the server;

obtaining the rules for each group from the server;

obtaining user-generated content generated on the client device by a user of the client device;

extracting language characteristics from the user-generated content created on the client device;

assigning one or more language characteristic scores to the user of the client device;

using the rules to generate a message for the user based on the users having similar language characteristic score(s) and group definition(s); and

displaying the message to the user on the client device.

13. A computer system for identifying message content to send to users based on the users' language characteristics, the system comprising:

a processor;

a memory coupled to the processor, wherein the memory stores instructions that, when executed by the processor, causes the system to perform the operations of:

obtaining user-generated content for a plurality of users, wherein each content item is associated with a user ID that uniquely identifies the user that generated the content item;

extracting language characteristics from the user-generated content;

assigning one or more language characteristic scores to each of the users;

clustering the users into groups using the language characteristic scores;

testing the responsiveness of each group to different message content by sending a plurality of test messages with different message content to at least a subset of users within each group and measuring the response rates to the test messages;

associating different message content with different language characteristic scores by identifying a message content to which each of the groups is most responsive and associating the message content with the group;

obtaining user-generated content for a new user;

extracting language characteristics from the new user's user-generated content;

assigning one or more language characteristic scores to the new user; and

sending a message to the new user with the message content previously associated with users having similar language characteristic score(s).

14. One or more non-transitory computer-readable media comprising computer program code that, when executed by a computer system, enables the computer system to perform the following method for identifying message content to send to users based on the users' language characteristics, the method comprising:

obtaining user-generated content for a plurality of users, wherein each content item is associated with a user ID that uniquely identifies the user that generated the content item;

extracting language characteristics from the user-generated content;

assigning one or more language characteristic scores to each of the users;

clustering the users into groups using the language characteristic scores;

testing the responsiveness of each group to different message content by sending a plurality of test messages with different message content to at least a subset of users within each group and measuring the response rates to the test messages;

associating different message content with different language characteristic scores by identifying a message content to which each of the groups is most responsive and associating the message content with the group;

obtaining user-generated content for a new user;

extracting language characteristics from the new user's user-generated content;

assigning one or more language characteristic scores to the new user; and

sending a message to the new user with the message content previously associated with users having similar language characteristic score(s).

15. The one or more non-transitory computer-readable media of claim 14 , wherein testing the responsiveness of each group to different message content comprises testing the responsiveness of each group to offers of different products to identify a best product for each group.

16. The one or more non-transitory computer-readable media of claim 15 , wherein, after a best product for each group is identified, testing the responsiveness of each group to certain vocabularies by sending test messages to at least a subset of users in each group that include an offer for the best product for the group and the certain vocabularies.

17. The one or more non-transitory computer-readable media of claim 16 , wherein, for each group, the message content associated with the group comprises the best product for the group and the vocabulary to which the group is most responsive.

18. The one or more non-transitory computer-readable media of claim 14 , wherein associating a message content with a group comprises associating a certain vocabulary with the group.

19. The one or more non-transitory computer-readable media of claim 14 , wherein associating a message content with the group comprises associating certain vocabulary rules with the group.

20. The one or more non-transitory computer-readable media of claim 14 , wherein the language characteristics on which the users are scored include type of vocabulary used.

21. The one or more non-transitory computer-readable media of claim 20 , wherein the language characteristics on which the users are scored also include morphology characteristics of the language.

22. The one or more non-transitory computer-readable media of claim 21 , wherein the morphology characteristics are one of: punctuation, capitalization, spelling errors, grammar errors, and acronyms.

23. The one or more non-transitory computer-readable media of claim 14 , wherein the language characteristics on which the users are scored include use of emoticons.