Systems, devices, and methods for content selection
Disclosed are systems, methods, and computer-readable storage media to present content on an electronic display. In one aspect, a method includes identifying a first candidate content and a second candidate content for presentation on an electronic display, determining a first probability and a second probability that the first candidate content and the second candidate content respectively will elicit a particular type of input response, determining a first weight and a second weight based on the first probability and the second probability respectively, selecting either the first content or the second content based on the first weight and the second weight; and presenting the selected content on the electronic display.
1. A method comprising:
identifying a first content item and a second content item;
storing a historical database comprising a first column that stores a time of day in which a given content item of a plurality of content items was presented, a second column that stores a view time representing an amount of time the given content item was viewed, and a third column that identifies the given content item;
estimating, based on the historical database, a plurality of probabilities of different types of input responses that will be received from a user in response to presenting the first and second content items to the user, the plurality of probabilities being computed in accordance with an expression defined as:
P t C =number of responses of type/total presentations of C ; and
where P t represents a probability that a response of type/will be provided for particular content C; and
selecting, as a selected content item, either the first content item or the second content item based on bid amounts associated with a first entity and a second entity.
2. The method of claim 1 , further comprising:
causing transmission of a first of the plurality of probabilities of the different types of input responses to the first entity associated with the first content item;
causing transmission of a second of the plurality of probabilities to the second entity associated with the second content item;
determining that the first entity has changed a maximum bid amount in response to the first of the plurality of probabilities;
determining that the second entity failed to change a bid amount based on the second of the plurality of probabilities; and
presenting the selected content item on an electronic display.
3. The method of claim 1 , further comprising:
establishing a first session for the user based on first user authentication credentials, the plurality of probabilities comprising a first probability that the first content item will elicit a first type of the different types of input responses and another probability that the first content item will elicit a second type of the different types of input responses, the plurality of probabilities being estimated based on a time of day, season and month during which the first content item will be presented to the user.
4. The method of claim 2 , further comprising:
estimating the second of the plurality of probabilities that the second content item will elicit a given type of input response of the different types of input responses from the user, the plurality of probabilities obtained from a classifier trained based on the historical database of characteristics of users generating responses, characteristics of the plurality of content items, and characteristics of channels over which the plurality of content items were presented.
5. The method of claim 1 , further comprising determining a first factor associated with the first content item and a second factor associated with the second content item, and determining a first weight and a second weight based on the first factor and the second factor, respectively.
6. The method of claim 5 , wherein determining the first weight comprises multiplying the first factor and a first probability to obtain the first weight.
7. The method of claim 1 , wherein a classifier is trained to generate the plurality of probabilities based on a plurality of input parameters comprising a life time distribution channel swipe rate for a given content item, a distribution channel swipe rate for the given content item, a swipe rate for a month by a given user, a lifetime distribution channel skip rate for the given content item, a swipe count for a previous month for the given content item, a total number of swipes for the given content item, an amount of time the given user has viewed given content item within a 30 day period, a distribution channel skip rate for the given content item, a skip rate per month for the given user, a lifetime number of swipes for the given content item, a number of skips the given user has performed in a preview month, a number of times the given content item has been skipped by a plurality of users who viewed the given content item, a number of times the given user has viewed given content item in the 30 day period, a life time number of skips for the given content item, and an indication of whether the given content item is displayed over a particular channel and a time of day, day of week, or month in which the given content item is displayed.
8. The method of claim 1 , further comprising:
receiving input in response to presentation of the selected content item;
categorizing the received input as either a first type of input or a second type of input; and
updating a historical response database based on categorizing of the received input.
9. The method of claim 8 , further comprising incrementing a total number of impressions for the selected content item in the historical response database in response to presentation of the selected content item.
10. The method of claim 9 , wherein estimating the plurality of probabilities comprises determining the total number of impressions of the first content item and a number of responses to the first content item having the first type.
11. The method of claim 10 , further comprising estimating a first probability by dividing the number of responses by the total number of impressions.
12. The method of claim 11 , further comprising filtering the total number of impressions and the number of responses to those impressions and responses for the user having an age within a predetermined range.
13. The method of claim 1 , wherein a classifier is trained to generate the plurality of probabilities based on a plurality of input parameters comprising a distribution channel swipe rate for a given content item, a total number of swipes for the given content item, a distribution channel skip rate for the given content item, a skip rate per month for a given user, a number of times the given content item has been skipped by a plurality of users who viewed the given content item, and a number of times the given user has viewed given content item in a 30 day period.
14. The method of claim 1 , wherein the first content item facilitates a first type of user interaction, and wherein the second content item facilitates a second type of user interaction, the first type of user interaction comprising adding a friend relationship within a social network, the second type of user interaction comprising scheduling an autonomous vehicle to pick up the user at a location indicated by a device of the user.
15. A system comprising:
one or more electronic hardware processors;
an electronic hardware memory, operatively coupled to the one or more electronic hardware processors, and storing instructions that configure the one or more electronic hardware processors to perform operations comprising:
identifying a first content item and a second content item;
storing a historical database comprising a first column that stores a time of day in which a given content item of a plurality of content items was presented, a second column that stores a view time representing an amount of time the given content item was viewed, and a third column that identifies the given content item;
estimating, based on the historical database, a plurality of probabilities of different types of input responses that will be received from a user in response to presenting the first and second content items to the user, the plurality of probabilities being computed in accordance with an expression defined as:
P t C =number of responses of type/total presentations of C ; and
where P t represents a probability that a response of type/will be provided for particular content C; and
selecting, as a selected content item, either the first content item or the second content item based on bid amounts associated with a first entity and a second entity.
16. The system of claim 15 , wherein the operations further comprise:
establishing a first session for the user based on first user authentication credentials, the plurality of probabilities comprising a first probability that the first content item will elicit a first type of the different types of input responses and another probability that the first content item will elicit a second type of the different types of input responses, the plurality of probabilities being estimated based on a time of day, season and month during which the first content item will be presented to the user.
17. The system of claim 15 , wherein a classifier is trained to generate the plurality of probabilities based on a plurality of input parameters comprising a distribution channel swipe rate for a given content item, a total number of swipes for the given content item, a distribution channel skip rate for the given content item, a skip rate per month for a given user, a number of times the given content item has been skipped by a plurality of users who viewed the given content item, and a number of times the given user has viewed given content item in a 30 day period.
18. A non-transitory computer readable medium comprising instructions that when executed cause at least one hardware processor to perform operations comprising:
identifying a first content item and a second content item;
storing a historical database comprising a first column that stores a time of day in which a given content item of a plurality of content items was presented, a second column that stores a view time representing an amount of time the given content item was viewed, and a third column that identifies the given content item;
estimating, based on the historical database, a plurality of probabilities of different types of input responses that will be received from a user in response to presenting the first and second content items to the user, the plurality of probabilities being computed in accordance with an expression defined as:
P t C =number of responses of type/total presentations of C ; and
where P t represents a probability that a response of type/will be provided for particular content C; and
selecting, as a selected content item, either the first content item or the second content item based on bid amounts associated with a first entity and a second entity.
19. The non-transitory computer readable medium of claim 18 , wherein the operations further comprise:
establishing a first session for the user based on first user authentication credentials, the plurality of probabilities comprising a first probability that the first content item will elicit a first type of the different types of input responses and another probability that the first content item will elicit a second type of the different types of input responses, the plurality of probabilities being estimated based on a time of day, season and month during which the first content item will be presented to the user.
20. The non-transitory computer readable medium of claim 18 , wherein the operations further comprise:
establishing a first session for the user based on first user authentication credentials, the plurality of probabilities comprising a first probability that the first content item will elicit a first type of the different types of input responses and another probability that the first content item will elicit a second type of the different types of input responses, the plurality of probabilities being estimated based on a time of day, season and month during which the first content item will be presented to the user.