Recommendation systems and methods using interest correlation
View Patent ↗A search technology generates recommendations with minimal user data and participation, and provides better interpretation of user data, such as popularity, thus obtaining breadth and quality in recommendations. It is sensitive to the semantic content of natural language terms and lets users briefly describe the intended recipient (i.e., interests, eccentricities, previously successful gifts). Based on that input, the recommendation software system and method determines the meaning of the entered terms and creatively discover connections to gift recommendations from the vast array of possibilities. The user may then make a selection from these recommendations. The search/recommendation engine allows the user to find gifts through connections that are not limited to previously available information on the Internet. Thus, interests can be connected to buying behavior by relating terms to respective items.
1. A computer implemented method of recommending products and services comprising the steps of:
processing media content and a plurality of user generated profiles to extract keywords;
generating a corpus of keywords using the extracted keywords;
identifying which keywords from the corpus commonly co-occur in the user generated profiles and media content; and
expanding a search query with additional search terms related to the search query, where the additional search terms are determined using one or more of the identified co-occurring keywords and by determining a correlation index between one or more terms of the search query and one or more keywords from the corpus, wherein the additional search terms have a correlation index above a threshold.
2. The method of claim 1 wherein the search query is a request for a media recommendation.
3. The method of claim 2 wherein the media recommendation search query recommends at least one of: a movie, music, a book, a television show, an article, and an event.
4. The method of claim 1 further comprising:
identifying the media content as having been accessed via a given user's profile.
5. The method of claim 4 wherein the given user previously positively rated the media content.
6. The method of claim 1 further comprising:
discarding an extracted keyword from the corpus based upon a frequency in the corpus of the discarded keyword.
7. The method of claim 6 wherein the discarded keyword is identified using a term frequency-inverse document frequency weighting measure.
8. The method of claim 1 wherein the additional search terms are not in a profile of a user requesting the search query.
9. The method of claim 1 wherein processing a plurality of user generated profiles to extract keywords comprises:
processing one or more videos in a given user profile to extract at least one of: a keyword displayed in the one or more videos and an audible keyword from the one or more videos.
10. A computer system for recommending products and services, the computer system comprising:
a processor; and
a memory with computer code instructions stored thereon, the processor and the memory, with the computer code instructions being configured to cause the system to:
process media content and a plurality of user generated profiles to extract keywords;
generate a corpus of keywords using the extracted keywords;
identify which keywords from the corpus commonly co-occur in the user generated profiles and media content; and
expand a search query with additional search terms related to the search query, where the additional search terms are determined using one or more of the identified co-occurring keywords and by determining a correlation index between one or more terms of the search query and one or more keywords from the corpus, wherein the additional search terms have a correlation index above a threshold.
11. The system of claim 10 wherein the search query is a request for a media recommendation.
12. The system of claim 11 wherein the media recommendation search query recommends at least one of: a movie, music, a book, a television show, an article, and an event.
13. The system of claim 10 wherein the processor and the memory, with the computer code instructions, are further configured to cause the system to:
identify the media content as having been accessed via a given user's profile.
14. The system of claim 13 wherein the given user previously positively rated the media content.
15. The system of claim 10 wherein the processor and the memory, with the computer code instructions, are further configured to cause the system to:
discard an extracted keyword from the corpus based upon a frequency in the corpus of the discarded keyword.
16. The system of claim 15 wherein the discarded keyword is identified using a term frequency-inverse document frequency weighting measure.
17. The system of claim 10 wherein the additional search terms are not in a profile of a user requesting the search query.
18. The system of claim 10 wherein, in processing a plurality of user generated profiles to extract keywords, the processor and the memory, with the computer code instructions, are further configured to cause the system to:
process one or more videos in a given user profile to extract at least one of: a keyword displayed in the one or more videos and an audible keyword from the one or more videos.
19. A computer program product for recommending products and services, the computer program product executed by a server in communication across a network with one or more clients and comprising:
a non-transitory computer readable storage medium, the computer readable storage medium comprising program instructions which, when executed by a processor causes:
processing media content and a plurality of user generated profiles to extract keywords;
generating a corpus of keywords using the extracted keywords;
identifying which keywords from the corpus commonly co-occur in the user generated profiles and media content; and
expanding a search query with additional search terms related to the search query, where the additional search terms are determined using one or more of the identified co-occurring keywords and by determining a correlation index between one or more terms of the search query and one or more keywords from the corpus, wherein the additional search terms have a correlation index above a threshold.