Selection and provision of digital components during display of content
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for the selection, provision and display of one or more digital components during display of content. Methods can include identifying a plurality of digital components that can be presented on the client device. A maximum number of digital components that can be presented in a slot of a content and the time duration of the slot is determined. For each digital component a score is generated based on the duration, a position requirement and the number of times the digital component is available for provision within the slot is generated. A first set of digital component is selected based on the scores and provided to the client device.
1 . A computer-implemented method, comprising:
identifying, by a content distribution system, a plurality of digital components and a slot within a particular content at which one or more digital components can be provided;
receiving, from a client device, a request to present the particular content on the client device, the request comprising event data specifying features from at least one of (i) the client device or (ii) a server from which the particular content is being requested;
for each respective digital component in the plurality of digital components, generating, using one or more machine learning models and based on the event data, a score, wherein the one or more machine learning models are trained to generate the score based on a plurality of parameters associated with a class of the respective digital component and the event data from the request;
selecting, based on the scores for the plurality of digital components, a first combination of digital components for provision within the slot, the first combination of digital components having a highest total score among different combinations of digital components from the plurality of digital components, the highest total score representing a likelihood of the first combination of digital components being provided in the slot; and
providing the first combination of digital components within the slot during display of the particular content on the client device.
2 . The computer-implemented method of claim 1 , wherein a parameter from the plurality of parameters comprises a value associated with the class of the respective digital component.
3 . The computer-implemented method of claim 1 , wherein a parameter from the plurality of parameters is based on a contextual relevance of the respective digital component with respect to the particular content.
4 . The computer-implemented method of claim 1 , wherein the features in the event data comprise one or both of (i) a name, or (ii) a network location, for the server from which the particular content is being requested.
5 . The computer-implemented method of claim 1 , wherein the features in the event data comprise one or both of (i) a name, or (ii) a network location, for the client device.
6 . The computer-implemented method of claim 1 , wherein the score generated by the one or more machine learning models is based on the plurality of parameters comprising one or more of:
(i) a duration factor representing a number of times that the respective digital component is available for provision within the slot,
(ii) a priority value indicating an importance of providing the respective digital component to the client device, or
(iii) a satisfaction ratio that represents a number of digital components of a particular duration that have previously been provided for display to client devices relative to a total number of digital components that are available for provision to client devices.
7 . The computer-implemented method of claim 1 , wherein providing the first combination of digital components within the slot during display of the particular content on the client device further comprises generating data for displaying the particular content.
8 . The computer-implemented method of claim 7 , wherein the data for displaying the particular content comprises at least one of: (i) a resource identifier for the particular content or (ii) attributes of the slot.
9 . A system, comprising:
one or more data processing apparatus; and
at least one memory coupled to the one or more data processing apparatus and storing one or more programming instructions that, when executed by the one or more data processing apparatus, perform operations comprising:
identifying, by a content distribution system, a plurality of digital components and a slot within a particular content at which one or more digital components can be provided;
receiving, from a client device, a request to present the particular content on the client device, the request comprising event data specifying features from at least one of (i) the client device or (ii) a server from which the particular content is being requested;
for each respective digital component in the plurality of digital components, generating, using one or more machine learning models and based on the event data, a score, wherein the one or more machine learning models are trained to generate the score based on a plurality of parameters associated with a class of the respective digital component and the event data from the request;
selecting, based on the scores for the plurality of digital components, a first combination of digital components for provision within the slot, the first combination of digital components having a highest total score among different combinations of digital components from the plurality of digital components, the highest total score representing a likelihood of the first combination of digital components being provided in the slot; and
providing the first combination of digital components within the slot during display of the particular content on the client device.
10 . The system of claim 9 , wherein a parameter from the plurality of parameters comprises a value associated with the class of the respective digital component.
11 . The system of claim 9 , wherein a parameter from the plurality of parameters is based on a contextual relevance of the respective digital component with respect to the particular content.
12 . The system of claim 9 , wherein the features in the event data comprise one or both of (i) a name, or (ii) a network location, for the server from which the particular content is being requested.
13 . The system of claim 9 , wherein the features in the event data comprise one or both of (i) a name, or (ii) a network location, for the client device.
14 . The system of claim 9 , wherein the score generated by the one or more machine learning models is based on the plurality of parameters comprising one or more of:
(i) a duration factor representing a number of times that the respective digital component is available for provision within the slot,
(ii) a priority value indicating an importance of providing the respective digital component to the client device, or
(iii) a satisfaction ratio that represents a number of digital components of a particular duration that have previously been provided for display to client devices relative to a total number of digital components that are available for provision to client devices.
15 . The system of claim 9 , wherein providing the first combination of digital components within the slot during display of the particular content on the client device further comprises generating data for displaying the particular content.
16 . The system of claim 15 , wherein the data for displaying the particular content comprises at least one of: (i) a resource identifier for the particular content or (ii) attributes of the slot.
17 . A non-transitory computer readable medium storing instructions that, when executed by one or more data processing apparatus, cause the one or more data processing apparatus to perform operations comprising:
identifying, by a content distribution system, a plurality of digital components and a slot within a particular content at which one or more digital components can be provided;
receiving, from a client device, a request to present the particular content on the client device, the request comprising event data specifying features from at least one of (i) the client device or (ii) a server from which the particular content is being requested;
for each respective digital component in the plurality of digital components, generating, using one or more machine learning models and based on the event data, a score, wherein the one or more machine learning models are trained to generate the score based on a plurality of parameters associated with a class of the respective digital component and the event data from the request;
selecting, based on the scores for the plurality of digital components, a first combination of digital components for provision within the slot, the first combination of digital components having a highest total score among different combinations of digital components from the plurality of digital components, the highest total score representing a likelihood of the first combination of digital components being provided in the slot; and
providing the first combination of digital components within the slot during display of the particular content on the client device.
18 . The non-transitory computer readable medium of claim 17 , wherein a parameter from the plurality of parameters comprises a value associated with the class of the respective digital component.
19 . The non-transitory computer readable medium of claim 17 , wherein a parameter from the plurality of parameters is based on a contextual relevance of the respective digital component with respect to the particular content.
20 . The non-transitory computer readable medium of claim 17 , wherein the features in the event data comprise one or both of (i) a name, or (ii) a network location, for the server from which the particular content is being requested.