Media content search engine incorporating text content and user log mining
Text features corresponding to pieces of media content (e.g., images, audio, multimedia content, etc.) are extracted from media content sources. One or more text features (e.g., one or more words) for a piece of media content are extracted from text associated with the piece of media content and text feature vectors generated therefrom and used during subsequent searching. Additional low-level feature vectors may also be extracted from the piece of media content and used during the subsequent searching. Relevance feedback can also be received from a user(s) identifying the relevance of pieces of media content rendered to the user in response to his or her search request. The relevance feedback is logged and can be used in determining how to respond to subsequent search requests, such as by modifying feature vectors (e.g., text feature vectors) corresponding to the pieces of media content for which relevance feedback is received.
1. A computer-implemented method comprising:
receiving search criteria;
identifying, based at least in part on the search criteria, a piece of media content to be rendered;
receiving, after rendering of the piece of media content, user feedback regarding relevancy of the piece of media content;
weighting for another piece of media content, based on the user feedback, both a result of comparing a high-level query vector to a high-level feature vector of the other piece of media content and a result of comparing a low-level query vector to a low-level feature vector of the other piece of media content; and
combining the weighted result to determine whether to identify the other piece of media content for rendering.
2. The method of claim 1 , further comprising generating a new high-level query vector and a new low-level query vector based at least in part on the search criteria.
3. The method of claim 1 , further comprising:
generating a user space vector corresponding to the piece of media content; and
using the user space vector corresponding to the piece of media content to modify the high-level feature vector corresponding to the piece of media content.
4. A The method of claim 1 , further comprising altering a weighting of one or more elements in the feature vector based on the user feedback.
5. The method of claim 1 , wherein the high-level feature vector of the other piece of media content is a text feature vector.
6. One or more computer-readable memories containing a computer program that is executable by a processor to perform the method recited in claim 1 .