Attribute-Based Item Similarity Using Collaborative Filtering Techniques

• US 20090077081A1
• Filed: 09/19/2007
• Published: 03/19/2009
• Est. Priority Date: 09/19/2007
• Status: Abandoned Application

• Current Assignee: Oath Inc. (Verizon Communications Inc.)
• Original Assignee: Oath Inc. (Verizon Communications Inc.)
• Inventors: Wang, Chao, Bhaskar, Anand, Sarma, Joydeep Sen
• Application Number:
  US11/857,847
• Publication Number:
  US 20090077081A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/90335 Query processing
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A system and method to compute attribute-based similarity between items using collaborative filtering techniques are described. Events input by a user over a network are received, the events further including a plurality of items and associated item metadata. A similarity value is further computed between each pair of items within the plurality of stems based on corresponding attributes of each item stored within the item metadata. Finally, recommendations of the items to the user are presented based on the corresponding calculated similarity value.

Claims

1. A method comprising:

receiving events input by a user over a network, said events further comprising a plurality of items and associated item metadata; and

calculating a similarity value between each pair of items within said plurality of items based on corresponding attributes of each item stored within said item metadata.

2. The method according to claim 1, further comprising:

presenting recommendations of said items to said user based on said corresponding calculated similarity value.

3. The method according to claim 1, further comprising:

storing said events within a data storage device; and

storing said item metadata within said data storage device, said item metadata further comprising said corresponding attributes of said each item.

4. The method according to claim 1, wherein said calculating further comprises:

calculating an attribute distance parameter between attribute values pertaining to each attribute of said each item; and

calculating an item distance parameter between any pair of items of said plurality of items based on said respective attribute distance parameters to determine a degree of similarity between said items characterized by said similarity value.

5. The method according to claim 4, wherein calculating said attribute distance parameter further comprising:

converting said plurality of items and said associated item metadata into attribute-value pairs corresponding to said each item;

storing said attribute-value pairs within attribute fables of a data storage device in connection with said associated user; and

calculating said attribute distance parameter based on said stored attribute-value pairs using a cosine dot product function.

6. The method according to claim 5, wherein calculating said item distance parameter further comprising:

retrieving said respective attribute distance parameters from said attribute tables;

calculating said item distance parameter based on said retrieved attribute distance parameters; and

storing said item distance parameter within respective item tables of said data storage device.

7. A computer readable medium containing executable instructions, which, when executed in a processing system, cause said processing system to perform a method comprising:

receiving events input by a user over a network, said events further comprising a plurality of items and associated item metadata; and

calculating a similarity value between each pair of items within said plurality of items based on corresponding attributes of each item stored within said item metadata.

8. The computer readable medium according to claim 7, wherein said method further comprises:

presenting recommendations of said items to said user based on said corresponding calculated similarity value.

9. The computer readable medium according to claim 7, wherein said method further comprises:

storing said events within a data storage device; and

storing said item metadata within said data storage device, said item metadata further comprising said corresponding attributes of said each item.

10. The computer readable medium according to claim 7, wherein said calculating further comprises:

calculating an attribute distance parameter between attribute values pertaining to each attribute of said each item; and

calculating an item distance parameter between any pair of items of said plurality of items based on said respective attribute distance parameters to determine a degree of similarity between said items characterized by said similarity value.

11. The computer readable medium according to claim 10, wherein calculating said attribute distance parameter further comprising:

converting said plurality of items and said associated item metadata into attribute-value pairs corresponding to said each item;

storing said attribute-value pairs within attribute tables of a data storage device in connection with said associated user; and

calculating said attribute distance parameter based on said stored attribute-value pairs using a cosine dot product function.

12. The computer readable medium according to claim 11, wherein calculating said item distance parameter further comprising:

retrieving said respective attribute distance parameters from said attribute tables;

calculating said item distance parameter based on said retrieved attribute distance parameters; and

storing said item distance parameter within respective item tables of said data storage device.

13. A system comprising:

at least one web server to receive events input by a user over a network, said events further comprising a plurality of items and associated item metadata; and

a processing engine coupled to said at least one web server to compute a similarity value between each pair of items within said plurality of items based on corresponding attributes of each item stored within said item metadata.

14. The system according to claim 13, wherein said processing engine further presents recommendations of said items to said user based on said corresponding calculated similarity value.

15. The system according to claim 13, wherein said processing engine further stores said events within a data storage device and stores said item metadata within said data storage device, said item metadata further comprising said corresponding attributes of said each item.

16. The system according to claim 13, wherein said processing engine further calculates an attribute distance parameter between attribute values pertaining to each attribute of said each item, and calculates an item distance parameter between any pair of items of said plurality of items based on said respective attribute distance parameters to determine a degree of similarity between said items characterized by said similarity value.

17. The system according to claim 16, wherein said processing engine further converts said plurality of items and said associated item metadata into attribute-value pairs corresponding to said each item, stores said attribute-value pairs within attribute tables of a data storage device in connection with said associated user, and calculates said attribute distance parameter based on said stored attribute-value pairs using a cosine dot product function.

18. The system according to claim 17, wherein said processing engine further retrieves said respective attribute distance parameters from said attribute tables, calculates said item distance parameter based on said retrieved attribute distance parameters, and stores said item distance parameter within respective item tables of said data storage device.