Character based media analytics
Techniques for analyzing media content are described. One technique generally comprises performing a regression analysis for characters in a plurality of media content based on user demographics, content outcome measure, and character models. The technique determines an attribute of significance. In some embodiments, the technique selects media content for display that depicts a character having at least a threshold value of the attribute of significance. In some embodiments, the technique displays media analytics for the attribute of significance determined based on a value of the attribute of significance exceeding a threshold significance value.
1. A computer-implemented method comprising:
determining a first outcome measure value of a desired outcome measure for a first sample media content;
determining a second outcome measure value of the desired outcome measure for a second sample media content;
accessing first respective character models of one or more characters represented in the first sample media content;
determining a first attribute value of a first attribute of the one or more characters represented in the first sample media content based on the first respective character models;
determining a second attribute value of a second attribute of the one or more characters represented in the first sample media content based on the first respective character models;
accessing second respective character models of one or more characters represented in the second sample media content;
determining a third attribute value of the first attribute of the one or more characters represented in the second sample media content based on the second respective character models;
determining a fourth attribute value of the second attribute of the one or more characters represented in the second sample media content based on the second respective character models;
performing a regression using the first outcome measure value, the second outcome measure value, the first attribute value, the second attribute value, the third attribute value, and the fourth attribute value, wherein performing the regression comprises estimating a statistical relationship between the desired outcome measure and the first attribute and a statistical relationship between the desired outcome measure and the second attribute;
determining, based on the regression, that the first attribute or the second attribute is an attribute of significance for the desired outcome measure;
determining that a character in a media content has at least a threshold value of the attribute of significance;
based on the character in the media content having at least the threshold value of the attribute of significance, selecting the media content from among a plurality of media content; and
providing the selected media content for display.
2. The computer-implemented method of claim 1 , further comprising determining a target demographic,
wherein determining that the first attribute or the second attribute is an attribute of significance comprises determining that the first attribute is an attribute of significance corresponding to the desired outcome measure and the target demographic, and
wherein providing the selected media content for display comprises providing the selected media content for display to a user based on the user having the target demographic.
3. The computer-implemented method of claim 1 , wherein determining that the first attribute or the second attribute is the attribute of significance comprises determining that the first attribute is the attribute of significance based on a regression value of the first attribute exceeding a threshold significance value.
4. The computer-implemented method of claim 1 , wherein the desired outcome measure is a click-through rate.
5. The computer-implemented method of claim 1 , wherein the desired outcome measure is a playback completion rate indicative of whether an interaction lasted for at least a threshold period of time.
6. The computer-implemented method of claim 1 , wherein the attribute of significance refers to a character quality, and wherein the character quality is a career, a demographic, a location, a social trait, a physical trait, or an intellectual trait.
7. The computer-implemented method of claim 1 , wherein the plurality of media content is a plurality of advertisements.
8. A non-transitory computer-readable medium having stored therein instructions that upon execution by a processor, cause performance of a set of acts comprising:
determining a first outcome measure value of a desired outcome measure for a first sample media content;
determining a second outcome measure value of the desired outcome measure for a second sample media content;
accessing first respective character models of one or more characters represented in the first sample media content;
determining a first attribute value of a first attribute of the one or more characters represented in the first sample media content based on the first respective character models;
determining a second attribute value of a second attribute of the one or more characters represented in the first sample media content based on the first respective character models;
accessing second respective character models of one or more characters represented in the second sample media content;
determining a third attribute value of the first attribute of the one or more characters represented in the second sample media content based on the second respective character models;
determining a fourth attribute value of the second attribute of the one or more characters represented in the second sample media content based on the second respective character models;
performing a regression using the first outcome measure value, the second outcome measure value, the first attribute value, the second attribute value, the third attribute value, and the fourth attribute value, wherein performing the regression comprises estimating a statistical relationship between the desired outcome measure and the first attribute and a statistical relationship between the desired outcome measure and the second attribute;
determining, based on the regression, that the first attribute or the second attribute is an attribute of significance for the desired outcome measure;
determining that a character in a media content has at least a threshold value of the attribute of significance;
based on the character in the media content having at least the threshold value of the attribute of significance, selecting the media content from among a plurality of media content; and
providing the selected media content for display.
9. The non-transitory computer-readable medium of claim 8 , wherein the set of acts further comprises determining a target demographic,
wherein determining that the first attribute or the second attribute is an attribute of significance comprises determining that the first attribute is an attribute of significance corresponding to the desired outcome measure and the target demographic, and
wherein providing the selected media content for display comprises providing the selected media content for display to a user based on the user having the target demographic.
10. The non-transitory computer-readable medium of claim 8 , wherein determining that the first attribute or the second attribute is the attribute of significance comprises determining that the first attribute is the attribute of significance based on a regression value of the first attribute exceeding a threshold significance value.
11. The non-transitory computer-readable medium of claim 8 , wherein the plurality of media content is a plurality of advertisements.
12. A computing system comprising a processor coupled to a memory, the computing system configured for performing a set of acts comprising:
determining a first outcome measure value of a desired outcome measure for a first sample media content;
determining a second outcome measure value of the desired outcome measure for a second sample media content;
accessing first respective character models of one or more characters represented in the first sample media content;
determining a first attribute value of a first attribute of the one or more characters represented in the first sample media content based on the first respective character models;
determining a second attribute value of a second attribute of the one or more characters represented in the first sample media content based on the first respective character models;
accessing second respective character models of one or more characters represented in the second sample media content;
determining a third attribute value of the first attribute of the one or more characters represented in the second sample media content based on the first respective character models;
determining a fourth attribute value of the second attribute of the one or more characters represented in the second sample media content based on the first respective character models;
performing a regression using the first outcome measure value, the second outcome measure value, the first attribute value, the second attribute value, the third attribute value, and the fourth attribute value, wherein performing the regression comprises estimating a statistical relationship between the desired outcome measure and the first attribute and a statistical relationship between the desired outcome measure and the second attribute;
determining, based on the regression, that the first attribute or the second attribute is an attribute of significance for the desired outcome measure;
determining that a character in a media content has at least a threshold value of the attribute of significance;
based on the character in the media content having at least the threshold value of the attribute of significance, selecting the media content from among a plurality of media content; and
providing the selected media content for display.
13. The computing system of claim 12 , wherein the set of acts further comprises determining a target demographic,
wherein determining that the first attribute or the second attribute is an attribute of significance comprises determining that the first attribute is an attribute of significance corresponding to the desired outcome measure and the target demographic, and
wherein providing the selected media content for display comprises providing the selected media content for display to a user based on the user having the target demographic.
14. The computing system of claim 12 , wherein determining that the first attribute or the second attribute is the attribute of significance comprises determining that the first attribute is the attribute of significance based on a regression value of the first attribute exceeding a threshold significance value.