Interestingness ranking of media objects
Media objects, such as images or soundtracks, may be ranked according to a new class of metrics known as “interestingness.” These rankings may be based at least in part on the quantity of user-entered metadata concerning the media object, the number of users who have assigned metadata to the media object, access patterns related to the media object, and/or a lapse of time related to the media object.
1. An apparatus comprising:
a processor;
a memory;
logic stored on said memory and executed by said processor for detecting a user interaction with a posted media object posted on a website by a posting user, the media object comprising metadata assigned to the media object;
logic stored on said memory and executed by said processor for identifying a source of the interaction, the source comprising one of a search results page, a group pool of media objects, or an email message;
logic stored on said memory and executed by said processor for classifying the user interaction based on the access pattern, the classifying comprising:
classifying a type of the access pattern as a view if the source is identified as a group pool of media objects and the user interaction is a click of the posted media object,
classifying a type of the access pattern as a click through if the source is identified as a click of the posted media object displayed on a search results page, and
classifying a type of the access pattern as a view if the source is an email message;
logic stored on said memory and executed by said processor for selecting an access pattern weight based on the type of the access pattern and a plurality of metadata weights based on the type of metadata assigned to the media object, wherein the access pattern weight and each of the metadata weights comprise predefined numerical coefficients;
logic stored on said memory and executed by said processor for scoring the media object with an interestingness score, the scoring based in part on the access pattern weight, the metadata weights, and a relationship between the posting user and the user accessing the posted media object; and
logic stored on said memory and executed by said processor for retrieving a plurality of responsive images in response to a search query, the responsive images including the media object;
logic stored on said memory and executed by said processor for removing the media object from the responsive images if the interestingness score is below a pre-determined threshold; and
logic stored on said memory and executed by said processor for causing display of the media object on the website if the interestingness score is above the pre-determined threshold.
2. The apparatus of claim 1 , wherein the logic for scoring the media object further comprises logic for determining a relationship between the posting user and an accessing user.
3. The apparatus of claim 2 , further comprising logic for increasing the score associated with the media object based on an identified similarity of interests between the accessing user and the posting user.
4. The apparatus of claim 1 , wherein the scoring of the media object is based upon the identity of an accessing user.
5. The apparatus of claim 1 , wherein the user interaction is a number of viewings of the media object.
6. The apparatus of claim 1 , wherein the user interaction is a number of click throughs of the media object.
7. The apparatus of claim 1 , wherein the user interaction is entering a search query.
8. The apparatus of claim 1 , further comprising logic for determining a location associated with an accessing user.
9. The apparatus of claim 8 , further comprising logic for determining a location associated with another user having a predefined relationship with the accessing user.
10. The apparatus of claim 1 , further comprising logic for decrementing the score over time.
11. A method comprising:
detecting, by a computing device, a user interaction with a posted media object posted on a website by a posting user, the media object comprising metadata assigned to the media object;
identifying, by the computing device, a source of the interaction, the source comprising one of a search results page, a group pool of media objects, or an email message;
classifying, by the computing device, the user interaction based on the access pattern, the classifying comprising:
classifying a type of the access pattern as a view if the source is identified as a group pool of media objects and the user interaction is a click of the posted media object,
classifying a type of the access pattern as a click through if the source is identified as a selection of the posted media object displayed on a search results page, and
classifying a type of the access pattern as a view if the source is an email message;
selecting, by the computing device, an access pattern weight based on the type of the access pattern and a plurality of metadata weights based on the type of metadata assigned to the media object, wherein the access pattern weight and each of the metadata weights comprise predefined numerical coefficients;
scoring, by the computing device, the media object with an interestingness score, the scoring based in part on the access pattern weight, the metadata weights, and a relationship between the posting user and a user of the plurality of users accessing the posted media object, and the relationship between the location associated with the media object; and
retrieving, by the computing device, a plurality of responsive images in response to a search query, the responsive images including the media object;
removing, by the computing device, the media object from the responsive images if the interestingness score is below a pre-determined threshold; and
causing, by the computing device, display of the media object on the website to be adjusted based on the interestingness score if the interestingness score is above the pre-determined threshold.
12. The method of claim 11 , wherein the scoring of the media object further comprises determining a relationship between the posting user and an accessing user.
13. The method of claim 12 , further comprising increasing the score associated with the media object based on an identified similarity of interests between the accessing user and the posting user.
14. The method of claim 11 , wherein the scoring of the media object is based upon the identity of an accessing user.
15. The method of claim 11 , wherein the user interaction is a number of viewings of the media object.
16. The method of claim 11 , wherein the user interaction is a number of click throughs of the media object.
17. The method of claim 11 , wherein the user interaction is entering a search query.
18. The method of claim 11 , further comprising determining a location associated with an accessing user.
19. The method of claim 18 , further comprising determining a location associated with another user having a predefined relationship with the accessing user.
20. The method of claim 11 , further comprising decrementing the score over time.