Method and system of optimizing a ranked list of recommended items
First Claim
Patent Images
1. A method of optimizing an output ranked list of recommended items given an input user, an input item list, and an input context, comprising:
- providing a multidimensional data set that comprises information of interactions from a plurality of users with a plurality of items and in a plurality of contexts;
factorizing the multidimensional data set into a number of two-dimensional matrices, the number of two-dimensional matrices being equivalent to the number of dimensions that the multidimensional data set has;
computing a mathematical recommendation model by optimizing an objective function over the two-dimensional matrices into which the multidimensional data set has been factorized, the recommendation model comprising a score value for each combination of user, item and context;
and computing the output ranked list by applying the computed recommendation model to the input user, input item list and input context,wherein the recommendation model further comprises a ranked list of recommended items for each user and context, being each ranked list determined by sorting the scores of the plurality of items for each user and context; and
wherein the objective function is a continuous function with infinite continuous derivatives that quantifies a relevance of the recommended items of each ranked list of the recommendation model, calculated over at least some of the plurality of users and over at least some of the plurality of contexts.
1 Assignment
0 Petitions
Accused Products
Abstract
A method and system of optimizing a ranked list (5) of recommended items that is based in a multidimensional data set (2) comprising context-aware information about the of a plurality of users and a plurality of items. A mathematical recommendation model (3) is trained with the multidimensional data set (2) by applying a smooth objective function that allows the use of fast optimizing algorithm and that quantifies the relevance of the ranked lists provided by an optimization algorithm.
-
Citations
18 Claims
-
1. A method of optimizing an output ranked list of recommended items given an input user, an input item list, and an input context, comprising:
-
providing a multidimensional data set that comprises information of interactions from a plurality of users with a plurality of items and in a plurality of contexts; factorizing the multidimensional data set into a number of two-dimensional matrices, the number of two-dimensional matrices being equivalent to the number of dimensions that the multidimensional data set has; computing a mathematical recommendation model by optimizing an objective function over the two-dimensional matrices into which the multidimensional data set has been factorized, the recommendation model comprising a score value for each combination of user, item and context; and computing the output ranked list by applying the computed recommendation model to the input user, input item list and input context, wherein the recommendation model further comprises a ranked list of recommended items for each user and context, being each ranked list determined by sorting the scores of the plurality of items for each user and context; and
wherein the objective function is a continuous function with infinite continuous derivatives that quantifies a relevance of the recommended items of each ranked list of the recommendation model, calculated over at least some of the plurality of users and over at least some of the plurality of contexts.- View Dependent Claims (4, 5, 6, 7, 8, 9, 10, 11, 15, 16, 17, 18)
-
-
2. A system of optimizing an output ranked list of recommended items given an input user and an input item list comprising context awareness means adapted to determine an input context of the user;
- an interface adapted to show information and receive commands from the input user; and
computing means adapted to;provide a multidimensional data set that comprises information of interactions from a plurality of users with a plurality of items and in a plurality of contexts; factorize the multidimensional data set into a number of two-dimensional matrices, the number of two-dimensional matrices being equivalent to the number of dimensions that the multidimensional data set has; compute a mathematical recommendation model by optimizing an objective function over the two-dimensional matrices into which the multidimensional data set has been factorized, the recommendation model comprising a score value for each combination of user, item and context; and compute the output ranked list by applying the computed recommendation model to the input user, input item list and input context, wherein the recommendation model further comprises a ranked list of recommended items for each user and context, being each ranked list determined by sorting the scores of the plurality of items for each user and context; and
in that the objective function is a continuous function with infinite continuous derivatives that quantifies a relevance of the recommended items of each ranked list of the recommendation model, calculated over at least some of the plurality of users and over at least some of the plurality of contexts.- View Dependent Claims (12, 13)
- an interface adapted to show information and receive commands from the input user; and
-
3. A computer program comprising computer program code means adapted to optimize an output ranked list of recommended items by:
-
providing a multidimensional data set that comprises information of interactions from a plurality of users with a plurality of items and in a plurality of contexts; factorizing the multidimensional data set into a number of two-dimensional matrices, the number of two-dimensional matrices being equivalent to the number of dimensions that the multidimensional data set has; computing a mathematical recommendation model by optimizing an objective function over the two-dimensional matrices into which the multidimensional data set has been factorized, the recommendation model comprising a score value for each combination of user, item and context; and computing the output ranked list by applying the computed recommendation model to the input user, input item list and input context, wherein the recommendation model further comprises a ranked list of recommended items for each user and context, being each ranked list determined by sorting the scores of the plurality of items for each user and context;
wherein the objective function is a continuous function with infinite continuous derivatives that quantifies a relevance of the recommended items of each ranked list of the recommendation model, calculated over at least some of the plurality of users and over at least some of the plurality of contexts; andwherein said program is run on a computer, a digital signal processor, a field-programmable gate array, an application-specific integrated circuit, a micro-processor, a micro-controller, or any other form of programmable hardware. - View Dependent Claims (14)
-
Specification