System and method for an electronic product advisor
First Claim
1. A computer-implemented method for product recommendation, the method comprising:
- by at least one computer system,determining a similarity percentage between a first list of products and each of a plurality of a second list of products, wherein each product that appears on at least one second list and does not appear on the first list is a candidate product;
for each candidate product,determining the sum, across the second lists, of each instance of the candidate product weighted by the determined similarity percentage for the second list upon which the instance of the candidate product appears; and
recommending candidate products in order of determined sum of the candidate product.
4 Assignments
0 Petitions
Accused Products
Abstract
A system and method operates on a client device and acquires a suspect list of user products based on information derived from the client device. The system normalizes the list, and the user confirms the accuracy of the product list. The user product list is sent to a server where the user product list is compared to other lists using collaborative filtering techniques. The collaborative filtering techniques determine products of interest for the use and the level of interest of the user. The system computes a similarity measure based upon the number of similar products that match the user'"'"'s product list and rankings provided by the user and others. Demographic and behavioral data may also be used in performing the comparison and the similarity measure. The system acquires editorial rankings of products from other users and provides a ranked list of recommended products based upon the editorial rankings.
144 Citations
22 Claims
-
1. A computer-implemented method for product recommendation, the method comprising:
by at least one computer system, determining a similarity percentage between a first list of products and each of a plurality of a second list of products, wherein each product that appears on at least one second list and does not appear on the first list is a candidate product; for each candidate product, determining the sum, across the second lists, of each instance of the candidate product weighted by the determined similarity percentage for the second list upon which the instance of the candidate product appears; and recommending candidate products in order of determined sum of the candidate product. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
8. A computer program product for product recommendation, the computer program product comprising:
-
non-transmission media memory; and instructions, stored on the non-transmission media memory, that when executed by at least one processor; determine a similarity percentage between a first list of products and each of a plurality of a second list of products, wherein each product that appears on at least one second list and does not appear on the first list is a candidate product; for each candidate product, determine the sum, across the second lists, of each instance of the candidate product weighted by the determined similarity percentage for the second list upon which the instance of the candidate product appears; and recommend candidate products in order of determined sum of the candidate product. - View Dependent Claims (9, 10, 11, 12, 13, 14)
-
-
15. A system for product recommendation, the system comprising:
-
at least one processor; and instructions that when executed by the at least one processor; determine a similarity percentage between a first list of products and each of a plurality of a second list of products, wherein each product that appears on at least one second list and does not appear on the first list is a candidate product; for each candidate product, determine the sum, across the second lists, of each instance of the candidate product weighted by the determined similarity percentage for the second list upon which the instance of the candidate product appears; and recommend candidate products in order of determined sum of the candidate product. - View Dependent Claims (16, 17, 18, 19, 20, 21)
-
-
22. A computer-implemented method for product recommendation, the method comprising:
-
by at least one computer system, determining a similarity percentage between a first set of products and each of a plurality of a second set of products, wherein each product in the at least one second set and not in the first set is a candidate product; for each candidate product, determining the sum, across the second sets, of each instance of the candidate product weighted by the determined similarity percentage for the second set in which the instance of the candidate product appears; and recommending candidate products in order of determined sum of the candidate product.
-
Specification