Machine learning-based media content sequencing and placement
View Patent ↗A system performs an automated analysis on a set of related media content items, such as static and video display advertisements for a coordinated advertising campaign. The analysis can include, for example, recognition of products, services, brands, objects, music, speech, motion, colors and moods, in order to determine content profiles for the content items. Different sequences of the media content items are placed within the web browsing paths of individual users, and the responses to the sequences are monitored with respect to desired outcomes, such as the purchase of a product or the visiting of an advertiser's website. The content profiles, the sequences, the placements, and the responses are provided as input into a machine learning system that is trained to select sequences and placements of media content items that achieve the desired outcomes. The system can be trained in part or wholly using feedback from its own output.
1. A method performed by a computer system having at least one processor and a memory, the method comprising, the computer system:
for each media content item of a plurality of media content items, generating a content profile for the media content item by performing an automated analysis of the media content item, wherein the content profile comprises a plurality of attribute fields each of which attribute fields specifies a characteristic of the media content item determined based on the automated analysis;
for each sequence of a plurality of different ordered sequences of media content items selected from the plurality of media content items, and with respect to the each sequence, for each user of a plurality of individual users:
presenting an instance of the each sequence by, for each media content item in the each sequence, causing an instance of the each media content item to be presented to the each user in order as targeted web advertising placements within the each user's web browsing path at unrelated locations on the World Wide Web, and
for each instance in which a media content item in the each sequence is presented to the each user, associating the each instance in which the media content item is presented with:
the instance of the each sequence in which the media content item is presented, and
an interaction profile indicative of user responses related to the presentation of the media content item;
selecting a target media content item from the plurality of media content items to present to a target user based on:
a sequence specifying an ordering of media content items previously presented to the target user,
the content profiles for the plurality of media content items,
the interaction profiles for the instances in which media content items are presented, and
the associations between instances of media content items being presented and instances of sequences in which media content items are presented; and
causing the target media content item to be presented to the target user,
wherein one of the media content items comprises a video segment,
wherein the content profile for the one media content item comprises an inflection point specifying a time location within the video segment at which an event identified by the automated analysis occurs,
wherein the interaction profile for an instance in which the one media content item is presented comprises an input time location within the video segment at which the each user provided input,
wherein the interaction profile comprises an interaction score indicative of engagement of a viewing user with the media content item, and
wherein the interaction score for the instance in which the one media content item is presented is based on a comparison between the inflection point and the input time.
2. The method of claim 1 ,
wherein for the each instance in which a media content item in the each sequence is presented to the each user, further associating the each instance with:
a placement profile descriptive of a context in which the media content item is presented,
the method further comprising:
establishing a target placement profile for the target media content item based on:
the placement profiles for the instances in which media content items are presented, and
the interaction profiles for the instances in which media content items are presented,
wherein the target media content item is presented to the target user in accordance with the target placement profile.
3. The method of claim 1 , wherein the plurality of users consists of individual users tracked and targeted for web advertising placements by an advertising network, and wherein the placement profile includes a domain of a hosting web page in the context of which the media content item is presented.
4. The method of claim 1 , wherein a single interaction profile is associated with a plurality of instances in which the media content item is presented, and wherein the single interaction profile is based on aggregated user response data associated with the plurality of instances.
5. The method of claim 4 , wherein the aggregated user response data is collected based on user actions performed subsequent to the plurality of instances in which the media content item is presented.
6. The method of claim 5 , wherein the aggregated user response data is based on product purchase data collected in association with one or more of a geographic area, a temporal duration, and viewing user geolocation tracking correlated with the plurality of instances in which the media content item is presented.
7. The method of claim 6 , wherein the product purchase data is indicative of one or more of online product purchases and in-person product purchases.
8. The method of claim 1 , wherein the interaction profile is based on active user input selected from the group consisting of: keyboard input, pointer selection, button press, voice capture, and touch screen selection.
9. The method of claim 1 , wherein each of the media content items comprises a video segment, and wherein the interaction profile is based on passive user input selected from the group consisting of: visibility of the presentation of the video segment on a computer display, absence of user input interrupting the presentation of the video segment, an indication that the presentation of the video segment continued to completion, and use of a camera to determine whether the user is viewing the presentation of the video segment.
10. The method of claim 9 , wherein the interaction profile is based on visibility of the video segment on a computer display determined based on whether another application window or scrolling of a page partly or fully obscures display of the video segment.
11. The method of claim 10 , wherein the interaction profile comprises one or more interaction fields, and wherein one of the at least one interaction fields indicates a time location within the video segment at which time location a user selected a selectable link presented in association the video segment.
12. The method of claim 1 , wherein one of the at least one attribute fields is configured to store identifications of spoken words identified in the media content item by the automated analysis.
13. The method of claim 12 , wherein one of the at least one attribute fields is configured to store a category assigned to the media content item by the automated analysis, and wherein the category is selected from a taxonomy of products and services.
14. The method of claim 13 , wherein one of the at least one attribute fields is configured to store identifications of objects identified in the media content item by the automated analysis.
15. The method of claim 13 , wherein one of the at least one attribute fields is configured to store identifications of music identified in the media content item by the automated analysis.
16. The method of claim 13 , wherein one of the at least one attribute fields is configured to store identifications of motion identified in the media content item by the automated analysis.
17. The method of claim 1 , wherein the interaction profile for at least one instance in which a media content item is presented to the each user comprises an indication of intent by a user to take an action.
18. The method of claim 1 , wherein the target media content item is one of the plurality of media content items.
19. The method of claim 1 , wherein the target media content item is not one of the plurality of media content items.
20. The method of claim 1 , wherein for at least one instance in which a media content item is presented, the presentation of the media content item is performed by a separate device from the computer system.
21. The method of claim 1 , wherein the sequences of media content items comprise video segments and non-video display advertisements.