Methods and apparatus to generate recommendations based on attribute vectors
Abstract
Methods and apparatus are disclosed to generate a recommendation, including an attribute vector aggregator to form a resultant attribute vector based on an input set of attribute vectors, the set of attribute vectors containing at least one of a media attribute vector, an attendee attribute vector, an artist attribute vector, an event attribute vector, or a venue attribute vector, and a recommendation generator, the recommendation generator including: a vector comparator to perform a comparison between an input attribute vector and other attribute vectors and a recommendation compiler to create one or more recommendations of at least one of media, an artist, an event, or a venue based on the comparison.
Claims
exact text as granted — not AI-modified1 - 28 . (canceled)
29 . An apparatus comprising:
one or more memories; instructions; and processor circuitry to execute the instructions to at least:
determine (a) a first result vector based on a first comparison of a query attribute vector and a first attribute vector and (b) a second result vector based on a second comparison of the query attribute vector and a second attribute vector, the query attribute vector based on one or more attribute vectors associated with a media player of a device;
apply (c) a first weight vector to the first result vector to determine a first weighted result vector and (d) a second weight vector to the second result vector to determine a second weighted result vector, the first weight vector corresponding to the first attribute vector, the second weight vector corresponding to the second attribute vector;
determine a first scalar value representative of the first weighted result vector and a second scalar value representative of the second weighted result vector; and
cause transmission of a recommendation associated with media to the device, the recommendation based on ordering of the first scalar value or the second scalar value.
30 . The apparatus of claim 29 , wherein the processor circuitry is to generate the first weight vector and the second weight vector, the first weight vector and the second weight vector based on a trend associated with the one or more attribute vectors associated with the media player of the device.
31 . The apparatus of claim 29 , wherein the processor circuitry is to:
access descriptive information associated with the media player of the device; access a third attribute vector and a fourth attribute vector based on the descriptive information; and aggregate the third attribute vector and the fourth attribute vector to generate the query attribute vector.
32 . The apparatus of claim 31 , wherein to aggregate the third attribute vector and the fourth attribute vector, the processor circuitry is to:
assign a first scalar weight to the third attribute vector to determine a third weighted attribute vector; apply a second scalar weight to the fourth attribute vector to determine a fourth weighted attribute vector; and determine a weighted average of the third weighted attribute vector and the fourth weighted attribute vector.
33 . The apparatus of claim 29 , wherein the recommendation associated with the media is based on a lower one of the first scalar value or the second scalar value.
34 . The apparatus of claim 29 , wherein the recommendation associated with the media includes a playlist that identifies a first media sample associated with the first attribute vector and a second media sample associated with the second attribute vector.
35 . The apparatus of claim 29 , wherein the recommendation associated with the media identifies at least one a song, an artist, an event, a playlist, or a venue.
36 . At least one non-transitory computer readable medium comprising instructions that, when executed, cause processor circuitry to at least:
determine (a) a first result vector based on a first comparison of a query attribute vector and a first attribute vector and (b) a second result vector based on a second comparison of the query attribute vector and a second attribute vector, the query attribute vector based on one or more attribute vectors associated with a media player of a device; apply (c) a first weight vector to the first result vector to determine a first weighted result vector and (d) a second weight vector to the second result vector to determine a second weighted result vector, the first weight vector corresponding to the first attribute vector, the second weight vector corresponding to the second attribute vector; determine a first scalar value representative of the first weighted result vector and a second scalar value representative of the second weighted result vector; and cause transmission of a recommendation associated with media to the device, the recommendation based on ordering of the first scalar value or the second scalar value.
37 . The at least one non-transitory computer readable medium of claim 36 , wherein the instructions, when executed, cause the processor circuitry generate the first weight vector and the second weight vector, the first weight vector and the second weight vector based on a trend associated with the one or more attribute vectors associated with the media player of the device.
38 . The at least one non-transitory computer readable medium of claim 36 , wherein the instructions, when executed, cause the processor circuitry:
access descriptive information associated with the media player of the device; access a third attribute vector and a fourth attribute vector based on the descriptive information; and aggregate the third attribute vector and the fourth attribute vector to generate the query attribute vector.
39 . The at least one non-transitory computer readable medium of claim 38 , wherein to aggregate the third attribute vector and the fourth attribute vector, the instructions, when executed, cause the processor circuitry:
assign a first scalar weight to the third attribute vector to determine a third weighted attribute vector; apply a second scalar weight to the fourth attribute vector to determine a fourth weighted attribute vector; and determine a weighted average of the third weighted attribute vector and the fourth weighted attribute vector.
40 . The at least one non-transitory computer readable medium of claim 36 , wherein the recommendation associated with the media is based on a lower one of the first scalar value or the second scalar value.
41 . The at least one non-transitory computer readable medium of claim 36 , wherein the recommendation associated with the media includes a playlist that identifies a first media sample associated with the first attribute vector and a second media sample associated with the second attribute vector.
42 . The at least one non-transitory computer readable medium of claim 36 , wherein the recommendation associated with the media identifies at least one a song, an artist, an event, a playlist, or a venue.
43 . A method comprising:
determining (a) a first result vector based on a first comparison of a query attribute vector and a first attribute vector and (b) a second result vector based on a second comparison of the query attribute vector and a second attribute vector, the query attribute vector based on one or more attribute vectors associated with a media player of a device; applying, by executing an instruction with processor circuitry, (c) a first weight vector to the first result vector to determine a first weighted result vector and (d) a second weight vector to the second result vector to determine a second weighted result vector, the first weight vector corresponding to the first attribute vector, the second weight vector corresponding to the second attribute vector; determining, by executing an instruction with the processor circuitry, a first scalar value representative of the first weighted result vector and a second scalar value representative of the second weighted result vector; and transmitting a recommendation associated with media to the device, the recommendation based on ordering of the first scalar value or the second scalar value.
44 . The method of claim 43 , further including generating the first weight vector and the second weight vector, the first weight vector and the second weight vector based on a trend associated with the one or more attribute vectors associated with the media player of the device.
45 . The method of claim 43 , further including:
accessing descriptive information associated with the media player of the device; accessing a third attribute vector and a fourth attribute vector based on the descriptive information; and aggregating the third attribute vector and the fourth attribute vector to generate the query attribute vector.
46 . The method of claim 45 , wherein aggregating the third attribute vector and the fourth attribute vector includes:
assigning a first scalar weight to the third attribute vector to determine a third weighted attribute vector; applying a second scalar weight to the fourth attribute vector to determine a fourth weighted attribute vector; and determining a weighted average of the third weighted attribute vector and the fourth weighted attribute vector.
47 . The method of claim 43 , wherein the recommendation associated with the media is based on a lower one of the first scalar value or the second scalar value.
48 . The method of claim 43 , wherein the recommendation associated with the media includes a playlist that identifies a first media sample associated with the first attribute vector and a second media sample associated with the second attribute vector.
49 . The method of claim 43 , wherein the recommendation associated with the media identifies at least one a song, an artist, an event, a playlist, or a venue.Join the waitlist — get patent alerts
Track US2023024574A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.