# US 2012/0253884 A1 · Methods and apparatus for modeling relationships at multiple scales in ratings estimation

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2012/0253884 A1
- Document ID: US20120253884A1
- Status: Published application
- Kind: A1
- Application number: 13/526,112
- Filed: 2012-06-18
- Published: 2012-10-04
- Earliest claimed: 2007-09-26
- Assignees: AT&T Intellectual Property 1, L.P. (VIA transfers/ name change from AT&T Labs, Inc.)
- Inventors: Yehuda Koren; Robert Bell; Christopher Volinsky
- Web page: https://patentgrep.com/patent/US20120253884A1

## CPC

- G06Q 30/0201: Market modelling; Market analysis; Collecting market data
- G06Q 30/02: Marketing; Price estimation or determination; Fundraising
- G06Q 30/0627: {by specifying product or service characteristics, e.g. product dimensions}
- G06Q 30/0631: {Recommending goods or services}
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

Systems and techniques for generating item ratings for a user in order to allow for recommendations of selected items for that user. A set of known ratings of different items for a plurality of users is collected and maintained, and these known ratings are used to estimate rating factors influencing ratings, including user and item factors. Initial user and item factors are estimated and new user and item factors are successively added, with the original rating factors being progressively shrunk so as to reduce their magnitude and their contribution to the rating estimation as successive factors are added. When an appropriate number of user and item factors has been estimated, the rating factors are used to estimate ratings of items for a user, and the estimated ratings are employed to generate recommendations for that user.

## Claims

### Claim 1

A method for estimating a rating of a first item by a first user, using a sparse matrix of known ratings κ={(u, i)|rui is known} wherein rui are ratings by users u for items i, comprising:
by a processor, iteratively pre-computing a matrix P containing a plurality of user factors for each user u and a matrix Q containing a plurality of item factors for each item i, by performing the following for each fth successive factor until a difference between factor-based estimates and known ratings κ is below a threshold value:
computing a residual res ui for each known rating rui as follows:
resui←rui−Σl=1f−1PuiQil;
adjusting the residual resui to provide shrinkage as follows:
res
ui
←
support
ui

res
ui
support
ui
+
α


f
,
wherein α is a constant and supportui is a support behind the known rating rui;
computing an fth user factor Puf for each user u:
P
uf
←
∑
i
:
(
u
,
i
)
∈
k

res
ui

Q
if
∑
i
:
(
u
,
i
)
∈
k

Q
if
2
;
computing an fth item factor Qif for each item i:
Q
if
←
∑
u
:
(
u
,
i
)
∈
k

res
ui

P
uf
∑
u
:
(
u
,
i
)
∈
k

P
uf
2
;
and
by the processor, estimating the rating of the first item by the first user, as a vector product of item factors Qif for the first item and user factors Puf for the first user.

### Claim 2 (depends on 1)

The method of claim 1, wherein the difference between factor-based estimates and known ratings κ is defined as an error function Err(P, Q):
Err

(
P
,
Q
)
=
∑
(
u
,
i
)

(
r
ui
-
p
u
T

q
i
)
2
,
wherein pu is a uth row of P and qi is an ith column of Q.

### Claim 3 (depends on 1)

The method of claim 1, wherein the items are movies and the users are movie distribution service users.

### Claim 4

The method of claim further comprising determining an optimal number of factors f so as to capture the most prominent features of the ratings data while leaving out relatively insignificant features.

### Claim 5 (depends on 1)

The method of claim 1, wherein the factors f are late features of the sparse matrix not requiring additional external information.

### Claim 6 (depends on 1)

The method of claim 1, wherein iteratively pre-computing the matrix P and the matrix Q is performed off-line.

### Claim 7 (depends on 6)

The method of claim 6, wherein iteratively pre-computing the matrix P and the matrix Q is performed upon receipt of new rating data.

### Claim 8 (depends on 1)

The method of claim 1, wherein the support supportui behind the known rating rui is a minimum of a number of ratings by a user u and a number of users that rated item i.

### Claim 9

A non-transitory computer-usable medium having computer readable instructions stored thereon for execution by a processor to perform a method for estimating a rating of a first item by a first user, using a sparse matrix of known ratings κ={(u, i)|rui is known} wherein rui are ratings by users u for items the method comprising:
iteratively pre-computing a matrix P containing a plurality of user factors for each user u and a matrix Q containing a plurality of item factors for each item i, by performing the following for each fth successive factor until a difference between factor-based estimates and known ratings κ is below a threshold value:
computing a residual resui for each known rating rui as follows:
resui←rui−Σl=1f−1PulQil;
adjusting the residual resui to provide shrinkage as follows:
res
ui
←
support
ui

res
ui
support
ui
+
α


f
,
wherein α is a constant and supportui is a support behind the known rating rui;
computing an fth user factor Puf for each user u:
P
uf
←
∑
i
:
(
u
,
i
)
∈
k

res
ui

Q
if
∑
i
:
(
u
,
i
)
∈
k

Q
if
2
;
and
computing any fth item factor Qif for each item i:
Q
if
←
∑
u
:
(
u
,
i
)
∈
k

res
ui

P
uf
∑
u
:
(
u
,
i
)
∈
k

P
uf
2
;
and
estimating the rating of the first item by he first user, as a vector product of item factors Qif for the first item and user factors Puf for the first user.

### Claim 10 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein the difference between factor-based estimates and known ratings κ is defined as an error function Err(P,Q):
Err

(
P
,
Q
)
=
∑
(
u
,
i
)

(
r
ui
-
p
u
T

q
i
)
2
,
wherein pu is a uth row of P and qi is an ith column of Q.

### Claim 11 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein the items are movies and the users are movie distribution service users.

### Claim 12 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein the method further comprises determining an optimal number of factors f so as to capture the most prominent features of the ratings data while leaving out relatively insignificant features.

### Claim 13 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein the factors f are latent features of the sparse matrix not requiring additional external information.

### Claim 14 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein iteratively pre-computing the matrix P and the matrix Q is performed off-line.

### Claim 15 (depends on 14)

The non-transitory computer-usable medium of claim 14, wherein iteratively pre-computing the matrix P and the matrix Q is performed upon receipt of new rating data.

### Claim 16 (depends on 9)

The non-transitory computer-usable medium of claim 9, wherein the support supportui behind the known rating rui is a minimum of a number of ratings by a user u and a number of users that rated item i.
