US2013325879A1PendingUtilityA1
Method and apparatus for recommending media programs based on correlated user feedback
Est. expiryOct 13, 2030(~4.2 yrs left)· nominal 20-yr term from priority
G06Q 30/0201G06F 16/44G06Q 30/0281G06F 17/30058
64
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
A method and apparatus for recommending a media program of a set of media programs to a user of a set of users is disclosed. The method and apparatus computes a measure w ij of the implied similarity of a first media program (i) and a second media program (j) that corrects for the popularity of the media programs, thus resulting in a more accurate indication of the relatedness of the media programs.
Claims
exact text as granted — not AI-modified1 - 15 . (canceled)
16 . A method comprising:
accepting, by a computing device, user data for a set of users for a plurality of media programs, wherein the user data represents a user's opinion of a media program in the plurality of media programs; computing, by the computing device, a first factor, the first factor being based on a first number of users having a similar user's opinion of a first media program and a second media program; computing, by the computing device, a second factor, the second factor being based on a second number of users having the similar user's opinion of the first media program and a third number of users having the similar user's opinion of the second media program; determining, by the computing device, a correction factor based on a popularity of the second media program; correcting, by the computing device, the first factor by the correction factor; computing, by the computing device, a similarity of the first media program and the second media program using the corrected first factor and the second factor; and recommending, by the computing device, a media program based on the computed similarity.
17 . The method of claim 16 , wherein the similar user's opinion comprises whether a user liked the first media program or the second media program.
18 . The method of claim 17 , wherein the first media program or the second media program in which the user liked was sent to the user through a video delivery service.
19 . The method of claim 16 , wherein determining the correction factor based on the popularity of the second media program comprises:
estimating a number of instances in which the first media program and the second media program include the similar user's opinion, but the first media program and the second media program are determined to be unrelated; and removing the number of instances from the first factor before computing the similarity.
20 . The method of claim 19 , wherein estimating the number of instances comprises:
determining if the similar user's opinion for the first media program is received after a release of the second media program for viewing; removing a chance that a user could provide the similar user's opinion for the second media program; and computing the popularity as a number of users that had the similar user's opinion for the second media program by a number of chances users could provide the similar user's opinion for the second media program.
21 . The method of claim 16 , wherein determining the correction factor comprises:
generating the popularity of the second media program; computing a difference between a number of media programs with the similar user's opinion by each individual user of the set of users after a release of the second media program for viewing and an indication that each individual user of the set of users has the similar user's opinion of the first media program after release of the second media program; generating, for each individual user of the set of users, a product of the indication that each user of the set of users likes the first media program, a difference between the number of media programs with the similar user's opinion by each individual user and the indication, and the popularity of the second media program; and summing the product of the indication over all individual users in the set of users.
22 . The method of claim 21 , wherein computing the correction factor is as follows:
∑
u
[
w
(
u
)
I
u
(
i
)
P
(
j
)
(
N
u
(
j
)
-
I
(
i
>
j
)
)
]
,
where w(u) is an optional weighting factor, I u (i)is the indication that a user of the set of users has the similar user's opinion of the first media program, P(j) is the popularity, N u (j)−I u (i>j) is the difference between the number of media programs with the similar user's opinion by each individual user and the indication.
23 . The method of claim 16 , wherein computing the first factor comprises:
computing a first sum computed across the set of users, the first sum being a sum that each individual user of the set of users has the similar user's opinion about both the first media program and the second media program.
24 . The method of claim 23 , wherein computing the second factor comprises:
computing a second sum across the set of users, the second sum being a sum that each individual user of the set of users has the similar user's opinion of the first media program; computing a third sum across the set of users, the third sum being a sum that each user of the set of users has the similar user's opinion of the second media program; and computing the second factor based on the second sum and the third sum.
25 . The method of claim 24 , wherein determining the second factor comprises computing a root of a product of the second sum and the third sum.
26 . The method of claim 24 , wherein computing the popularity comprises:
computing a fourth sum across the set of users, of a number of times each user of the set of users has the similar user's opinion of the second media program; and computing a fifth sum, across the set of users, of a number of media programs each user of the set of users has the similar user's opinion of after a release of the second media program.
27 . The method of claim 26 , wherein computing the popularity comprises:
computing a ratio of the fourth sum and the fifth sum.
28 . The method of claim 26 , wherein computing the similarity is based on the first sum, the second sum, the third sum, the fourth sum, and the fifth sum.
29 . The method of claim 26 , wherein:
the
first
sum
comprises
∑
u
w
(
u
)
I
u
(
i
,
j
)
,
the
second
sum
comprises
∑
u
w
(
u
)
I
u
(
i
)
,
the
third
sum
comprises
∑
u
w
(
u
)
I
u
(
j
)
,
the
correction
factor
comprises
∑
u
[
w
(
u
)
I
u
(
i
)
P
(
j
)
(
N
u
(
j
)
-
I
(
i
>
j
)
)
]
,
and
the
similarity
comprises
w
ij
=
∑
u
w
(
u
)
I
u
(
i
,
j
)
-
∑
u
[
w
(
u
)
I
u
(
i
)
P
(
j
)
(
N
u
(
j
)
-
I
u
(
i
>
j
)
)
]
∑
u
w
(
u
)
I
u
(
i
)
*
∑
u
w
(
u
)
I
u
(
j
)
,
where w(u) is an optional weighting factor, I u (i) is an indication that a user of the set of users has the similar user's opinion of the first media program, I u (i, j) is an indication that a user in the set of users has the similar user's opinion for the first media program and the second media program, P(j) is the popularity, N u (j)−I u (i>j) is a difference between a number of media programs with the similar user's opinion by each individual user of the set of users after a release of the second media program for viewing and an indication that each individual user of the set of users has the similar user's opinion of the first media program after release of the second media program, and I u (j) is an indication that a user of the set of users has the similar user's opinion of the second media program.
30 . The method of claim 16 , wherein computing the similarity comprises computing a ratio of the corrected first factor to the second factor.
31 . A non-transitory computer-readable storage medium containing instructions, that when executed, control a computer system to be configured for:
accepting user data for a set of users for a plurality of media programs, wherein the user data represents a user's opinion of a media program in the plurality of media programs; computing a first factor, the first factor being based on a first number of users having a similar user's opinion of a first media program and a second media program; computing a second factor, the second factor being based on a second number of users having the similar user's opinion of the first media program and a third number of users having the similar user's opinion of the second media program; determining a correction factor based on a popularity of the second media program; correcting the first factor by the correction factor; computing a similarity of the first media program and the second media program using the corrected first factor and the second factor; and recommending a media program based on the computed similarity.
32 . The apparatus of claim 31 , wherein determining the correction factor based on the popularity of the second media program comprises:
estimating a number of instances in which the first media program and the second media program include the similar user's opinion, but the first media program and the second media program are determined to be unrelated; and removing the number of instances from the first factor before computing the similarity.
33 . The apparatus of claim 32 , wherein estimating the number of instances comprises:
determining if the similar user's opinion for the first media program is received after a release of the second media program for viewing; removing a chance that a user could provide the similar user's opinion for the second media program; and computing the popularity as a number of users that had the similar user's opinion for the second media program by a number of chances users could provide the similar user's opinion for the second media program.
34 . The apparatus of claim 31 , wherein determining the correction factor comprises:
generating the popularity of the second media program; computing a difference between a number of media programs with the similar user's opinion by each individual user of the set of users after a release of the second media program for viewing and an indication that each individual user of the set of users has the similar user's opinion of the first media program after release of the second media program; generating, for each individual user of the set of users, a product of the indication that each user of the set of users likes the first media program, a difference between the number of media programs with the similar user's opinion by each individual user and the indication, and the popularity of the second media program; and summing the product of the indication over all individual users in the set of users.
35 . An apparatus comprising:
one or more computer processors; and a non-transitory computer-readable storage medium comprising instructions, that when executed, control the one or more computer processors to be configured for: accepting user data for a set of users for a plurality of media programs, wherein the user data represents a user's opinion of a media program in the plurality of media programs; computing a first factor, the first factor being based on a first number of users having a similar user's opinion of a first media program and a second media program; computing a second factor, the second factor being based on a second number of users having the similar user's opinion of the first media program and a third number of users having the similar user's opinion of the second media program; determining a correction factor based on a popularity of the second media program; correcting the first factor by the correction factor; computing a similarity of the first media program and the second media program using the corrected first factor and the second factor; and recommending a media program based on the computed similarity.Join the waitlist — get patent alerts
Track US2013325879A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.