System and methods for generating optimal post times for social networking sites
A system and methods are disclosed for determining the ideal times for a person, software client, or other entity to post a message to a social networking site. An ideal time is a time when the post will have a maximum impact, where impact is some measure of success as defined by the posting entity. To determine the ideal times, the posting patterns of individual users are aggregated in a weighted fashion, taking into consideration both the entity's desired impact and the likelihood that the user will be online and able to view the post within a specified time frame.
1. A system for generating an ideal post time for a social network site for a user, the system comprising:
a user database for storing captured information about followers that follow activity of the user on the social networking site;
a timestamp database for storing collected times at which the followers post to the social networking site; and
a computer for generating the ideal post time for the user, wherein the computer calculates the ideal post time by acquiring follower attributes, computes and assigns a score to each of the followers based on the follower attributes and the captured information in the user database, and generates an activity model for each of the followers based on the collected times in the timestamp database.
2. The system of claim 1 wherein the follower attributes are acquired from at least one of a social importance ranking service or a source of data having information about the followers.
3. The system of claim 2 wherein the computer generating the ideal post time smoothes the activity models by applying a smoothing algorithm.
4. A method for generating an ideal post time for a user of a social networking site, the method comprising:
acquiring attributes of followers that follow the activity of the user on the social networking site;
collecting timelines for each of the followers according to an available sequence of follower posts and times of those posts;
computing and assigning a score to each of the followers based on user interests and the follower attributes.
5. The system of claim 1 , wherein the computer generating the ideal post time further weights the activity models based on the score assigned to each of the followers.
6. The system of claim 5 , wherein the computer generating the ideal post time further aggregates the weighted activity models of the followers.
7. The system of claim 1 , wherein the score is based in part on a number of follower's followers.
8. The system of claim 7 , wherein the score is higher for a large number of follower's followers.
9. The system of claim 7 , wherein the score is lower for a small number of follower's followers.
10. The system of claim 1 , wherein the system generates a histogram indicating when a follower is online and likely to see a post.
11. The system of claim 10 , wherein said computer generating the ideal post time generates said activity model based on said histogram indicating when a follower is online and likely to see a post.
12. The method for generating an ideal post time for a user of a social networking site of claim 4 , the method further comprising:
generating activity models for each of the followers based on the timelines;
weighting the activity models based on the score assigned to each of the followers; and
aggregating the weighted activity models of the followers.
13. The method of claim 12 wherein the follower attributes are acquired from at least one of a social importance ranking service or a source of data having information about the followers.
14. The method of claim 13 wherein the computer generating the ideal post time smoothes the activity models by applying a smoothing algorithm.
15. The method of claim 12 , wherein the score is based in part on a number of follower's followers.
16. The method of claim 15 , wherein the score is higher for a large number of follower's followers.
17. The method of claim 15 , wherein the score is lower for a small number of follower's followers.
18. The method of claim 12 , wherein the system generates a histogram indicating when a follower is online and likely to see a post.
19. The method of claim 18 , wherein said computer generating the ideal post time generates said activity model based on said histogram indicating when a follower is online and likely to see a post.