Social content suggestions based on connections
A system and method for recommending content to a user in a social network, including: logging user activity for the user in the social network; categorizing the user activity across all the user's networks, wherein each category is assigned a score based on relevance to the user; assigning weights to the user activities; calculating a social index score as a function of the weighted user activity categories; logging user content into categories; scoring the user content; and generating a content social index by weighting the content scores.
1. A method for recommending content in a social network, comprising:
logging a set of actions of a user;
categorizing the set of actions into interest categories;
weighting the interest categories to generate weighted interest categories, the weighting comprising:
for a first interest category associated with a first subset of the set of actions:
identifying one or more types of actions associated with the first interest category, respective types of actions associated with an action weight, the identifying comprising:
identifying a first type of action corresponding to a search action, a second type of action corresponding to a subscribe action, and a third type of action corresponding to a click action, the first type of action, the second type of action and the third type of action associated with the first interest category, the first type of action having a first action weight, the second type of action having a second action weight, and the third type of action having a third action weight; and
assigning a first weight to the first interest category based on the one or more types of actions associated with the first interest category and an action weight associated with each of the one or more types of actions;
calculating an interest social index for the user as a function of the weighted interest categories; and
providing, on a social page associated with the social network, and based on the interest social index, a first indication of a first suggested topic and a first value corresponding to a likelihood of one or more users in the social network engaging the first suggested topic and a second indication of a second suggested topic and a second value corresponding to a likelihood of one or more users in the social network engaging the second suggested topic.
2. The method of claim 1 , comprising:
calculating an influencer social index based on the interest social index and one or more searches which led to a profile of a second user.
3. The method of claim 1 , the assigning comprising:
multiplying the first action weight by a first portion of the first subset associated with the first type of action to generate a first weighted component of the first interest category;
multiplying the second action weight by a second portion of the first subset associated with the second type of action to generate a second weighted component of the first interest category; and
multiplying the third action weight by a third portion of the first subset associated with the third type of action to generate a third weighted component of the first interest category.
4. The method of claim 3 , the assigning comprising:
combining the first weighted component with the second weighted component and the third weighted component to yield the first weight.
5. The method of claim 1 , the providing performed responsive to a selection of a prompt on the social page.
6. The method of claim 1 , comprising:
logging content created by the user; and
scoring the content.
7. The method of claim 6 , the scoring the content comprising:
scoring the content based on a number of entities linking to the content.
8. The method of claim 6 , the scoring the content comprising:
scoring the content based on a category to which the content relates.
9. The method of claim 6 , the scoring the content comprising:
scoring the content based on interest social indexes of one or more users viewing the content.
10. The method of claim 1 , comprising:
logging a second set of actions of a second user within the social network of the user;
categorizing the second set of actions into a second set of interest categories;
weighting the second set of interest categories to generate a second set of weighted interest categories, the weighting comprising:
for a second interest category associated with a second subset of the second set of actions:
identifying one or more types of actions associated with the second interest category, respective types of actions associated with an action weight; and
assigning a second weight to the second interest category based on the one or more types of actions associated with the second interest category and an action weight associated with each of the one or more types of actions; and
calculating an interest social index for the second user as a function of the weighted interest categories.
11. The method of claim 10 , comprising:
calculating an influencer social index for the user based on the interest social index for the second user.
12. The method of claim 11 , comprising:
recommending new content for the user to generate based on the influencer social index and the interest social index for the user.
13. The method of claim 12 , the recommending comprising:
recommending the new content based on a content social index, the content social index derived based on content created by the user.
14. An information processing system for recommending content in a social network, said information processing system comprising:
a processor; and
memory comprising computer-executable instructions that when executed by the processor perform a method, comprising:
for a user:
logging a set of actions;
categorizing the set of actions into interest categories;
weighting the interest categories to generate weighted interest categories, the weighting comprising:
for a first interest category associated with a first subset of the set of actions:
identifying one or more types of actions associated with the first interest category, respective types of actions associated with an action weight, the identifying comprising:
identifying a first type of action corresponding to a search action, a subscribe action or a click action, a second type of action, and a third type of action, the first type of action, the second type of action and the third type of action associated with the first interest category, the first type of action having a first action weight, the second type of action having a second action weight, and the third type of action having a third action weight; and
assigning a first weight to the first interest category based on the one or more types of actions associated with the first interest category and an action weight associated with each of the one or more types of actions; and
calculating an interest social index for the user as a function of the weighted interest categories;
for a second user within the social network of the user:
logging a second set of actions;
categorizing the second set of actions into a second set of interest categories;
weighting the second set of interest categories to generate a second set of weighted interest categories; and
calculating an interest social index for the second user as a function of the weighted interest categories; and
calculating an influencer social index for the user based on the interest social index for the second user.
15. The information processing system of claim 14 , the method comprising:
recommending new content for the user to generate based on the influencer social index and the interest social index for the user.
16. The information processing system of claim 14 , the logging a set of actions comprising:
logging the set of actions from across multiple online mediums.
17. The information processing system of claim 14 , the first type of action corresponding to the search action.
18. The information processing system of claim 14 , the first type of action corresponding to the subscribe action.
19. The information processing system of claim 14 , the first type of action corresponding to the click action.
20. A computer program product comprising a non-transitory computer readable storage medium with computer-executable instructions stored thereon, said computer-executable instructions comprising:
logging a set of actions of a user;
categorizing the set of actions into interest categories;
weighting the interest categories to generate weighted interest categories, the weighting comprising:
for a first interest category associated with a first subset of the set of actions:
identifying one or more types of actions associated with the first interest category, respective types of actions associated with an action weight, the identifying comprising:
identifying a first type of action, a second type of action, and a third type of action associated with the first interest category, the first type of action having a first action weight, the second type of action having a second action weight, and the third type of action having a third action weight; and
assigning a first weight to the first interest category based on the one or more types of actions associated with the first interest category and an action weight associated with each of the one or more types of actions;
calculating an interest social index for the user as a function of the weighted interest categories; and
providing, on a social page, and based on the interest social index, a first indication of a first suggested topic and a first value corresponding to a likelihood of one or more users engaging the first suggested topic and a second indication of a second suggested topic and a second value corresponding to a likelihood of one or more users engaging the second suggested topic.