System and method for predicting missing product ratings utilizing covariance matrix, mean vector and stochastic gradient descent

• US 8,595,089 B1
• Filed: 02/15/2010
• Issued: 11/26/2013
• Est. Priority Date: 02/15/2010
• Status: Active Grant

• Current Assignee: William John James Roberts
• Original Assignee: William John James Roberts
• Inventors: Roberts, William John James
• Primary Examiner(s): Jarrett, Scott L
• Application Number:
  US12/705,932
• Time in Patent Office: 1,380 Days
• Field of Search: 705/7.29, 705/14.49, 705/14.51, 705/26.1, 705/26.7, 455/69, 706/12, 706/46, 347/1
• US Class Current: 705/26.7
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

A product recommender system where product ratings from a plurality of users are represented as plurality of vectors is disclosed. The ratings vectors are represented by a mean vector and a covariance matrix. The mean vector and the covariance matrix are estimated from a data-set of known product ratings. Product ratings are predicted using the mean vector and the covariance matrix.

Claims

1. A method of predicting a product rating comprising steps of:

retaining a set of observed product ratings on a computer storage device;

accessing the set from the computer storage device;

calculating a mean vector, via a computer processor;

estimating a single covariance matrix from the set;

initializing the covariance matrix using the equation R=N<sup>−

1/2</sup>SN<sup>−

1/2</sup>, where S denotes an un-normalized sample covariance matrix and elements of the diagonal matrix N denote the number of times each product was rated, predicting a product rating using the mean vector and the covariance matrix when the product rating is absent from the set; and

providing the product rating to an end user.

2. The method of claim 1, wherein the mean vector is estimated using a stochastic gradient descent approach.

3. The method of claim 1, wherein the covariance matrix is estimated using a stochastic gradient descent approach.

4. The method of claim 1, wherein the prediction of product rating by the given user is accomplished using a minimum means squared error predictor.

5. The method of claim 1, wherein calculations are performed using basic linear algebra subroutines.

6. The method of claim 1, wherein the covariance matrix is estimated using an expectation maximization algorithm.

7. The method of claim 1, further comprising the step of continuously updating the estimation of the covariance matrix.

8. The method of claim 7, further comprising the step of checking for convergence and if there is convergence halting the continuous updating of the estimation of the covariance matrix and saving the estimated mean vector and covariance matrix.

9. The method of claim 1, further comprising the step of calculating a confidence rating from at least one portion of the covariance matrix.

10. A method of predicting a product rating comprising steps of:

retaining a set of observed product ratings on a computer storage device;

accessing the set from the computer storage device;

calculating a mean vector using a stochastic gradient descent approach, via a computer processor, corresponding to a mean of the set;

initializing a single covariance matrix using the equation R=N<sup>−

1/2</sup>SN<sup>−

1/2</sup> where S denotes an un-normalized sample covariance matrix and elements of the diagonal matrix N denote the number of times each product was rated, and

estimating the single covariance matrix from the set using a stochastic gradient descent approach;

predicting a product rating using the mean vector and the covariance matrix wherein the product rating is absent from the set; and

providing the product rating to an end user.

11. A method of predicting a product rating comprising steps of:

retaining a set of observed product ratings on a computer storage device;

accessing the set from the computer storage device;

estimating a covariance matrix from the set using the equation R=N<sup>−

1/2</sup>SN<sup>−

1/2</sup> where S denotes an un-normalized sample covariance matrix and elements of the diagonal matrix N denote the number of times each product was rated,

predicting a product rating using the covariance matrix when the product rating is absent from the set; and

providing the product rating to an end user.