Personalized cross session diversity
View Patent ↗A system and method for personalizing cross session diversity is disclosed. The system receives a member opportunity request. In response, the system generates a list of members in response to the received member opportunity request, wherein the list of members is determined based on member profile data stored at a social networking system. For each member in the generated list of members, the system generates a profile value score based on the stored member profile data. The system ranks the members of the generated list at least in part based on the generated profile value scores. The system then selects one or more members in the list of members based on the ranking of members in the generated list.
1. A method comprising:
receiving an activity feed request from a first member during a first session;
selecting a plurality of potential activity feed items based on data stored in a member profile associated with the first member;
for a respective potential activity feed item in the plurality of potential activity feed items:
determining past session item display data associated with the respective potential activity feed item,
generating a relevance score for the respective potential activity feed item based on the stored member profile data;
using the determined past session item display data, generating a discount factor for the respective potential activity feed item;
discounting the relevance score based on the determined discount factor; and
ordering the plurality of potential activity feed items based on the discounted relevance score of each potential activity feed item.
2. The method of claim 1 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining a source member associated with the respective potential activity feed item; and
using stored past session data, determining a number of past session item views associated with the source member.
3. The method of claim 1 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining an activity feed item type of the respective potential activity feed item;
using stored past session data, determining a number of past session item views associated with the activity feed type of the respective potential activity feed item.
4. The method of claim 2 , wherein the discount factor for a particular potential activity feed item is based on the number of past session item views associated with the source member.
5. The method of claim 1 , wherein the discount factor for a particular potential activity feed item is based on the number of past session item views associated with the activity feed type of the respective potential activity feed item.
6. The method of claim 2 , wherein the discount factor for a particular potential activity feed item is based on the social graph connection between the requesting member and the source member.
7. The method of claim 2 , wherein the discount factor for a particular potential activity feed item is based on member interaction with content associated with the source member of the potential activity feed item.
8. The method of claim 2 , wherein the discount factor is determined based on both the social graph relationship between the requesting member and the source member and the item type of the particular potential activity feed item.
9. A system comprising:
one or more processors;
memory; and
one or more programs stored in the memory, the one or more programs comprising instructions for:
receiving an activity feed request from a first member during a first session;
selecting a plurality of potential activity feed items based on member profile stored for the first member;
for a respective potential activity feed item in the plurality of potential activity feed items:
determining past sessions item display data associated with the respective potential activity feed item,
generating a relevance score for the respective potential activity feed item based on the stored member profile data;
using the determined past session item display data, generating a discount factor for the respective potential activity feed item;
discounting the relevance score based on the determined discount factor; and
ordering the plurality of potential activity feed items based on the discounted relevance score of each potential activity feed item.
10. The system of claim 9 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining a source member associated with the respective potential activity feed item; and
using stored past session data, determining a number of past session item views associated with the source member.
11. The system of claim 9 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining an activity feed item type of the respective potential activity feed item;
using stored past session data, determining a number of past session item views associated with the activity feed type of the respective potential activity feed item.
12. The system of claim 10 , wherein the discount factor for a particular potential activity feed item is based on the number of past session item views associated with the source member.
13. The system of claim 9 , wherein the discount factor for a particular potential activity feed item is based on the number of past session item views associated with the activity feed type of the respective potential activity feed item.
14. The system of claim 10 , wherein the discount factor for a particular potential activity feed item is based on the social graph connection between the requesting member and the source member.
15. The system of claim 10 , wherein the discount factor for a particular potential activity feed item is based on member interaction with content associated with the source member of the potential activity feed item.
16. The system of claim 10 , wherein the discount factor is determined based on both the social graph relationship between the requesting member and the source member and the item type of the particular potential activity feed item.
17. A non-transitory computer readable storage medium storing one or more programs for execution by one or more processors, the one or more programs comprising instructions for:
receiving an activity feed request from a first member during a first session;
selecting a plurality of potential activity feed items based on member profile stored for the first member;
for a respective potential activity feed item in the plurality of potential activity feed items:
determining past sessions item display data associated with the respective potential activity feed item,
generating a relevance score for the respective potential activity feed item based on the stored member profile data;
using the determined past session item display data, generating a discount factor for the respective potential activity feed item;
discounting the relevance score based on the determined discount factor; and
ordering the plurality of potential activity feed items based on the discounted relevance score of each potential activity feed item.
18. The non-transitory computer readable storage medium of claim 17 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining a source member associated with the respective potential activity feed item; and
using stored past session data, determining a number of past session item views associated with the source member.
19. The non-transitory computer readable storage medium of claim 17 , wherein determining past session item display data includes:
for a respective potential activity feed item, determining an activity feed item type of the respective potential activity feed item;
using stored past session data, determining a number of past session item views associated with the activity feed type of the respective potential activity feed item.
20. The non-transitory computer readable storage medium of claim 18 , wherein the discount factor for a particular potential activity feed item is based on the number of past session item views associated with the source member.