Predicting future insertion zone metadata

A non-transitory computer-readable medium having computer executable instructions stored thereon, which when executed by a computing device cause a system to perform operations, the operations comprising:

• obtaining, from data associated with the new video, categorical metadata relating to the new video, the categorical metadata comprising at least one categorical variable;
• obtaining, for a particular categorical variable of the at least one categorical variable, an associated insertion value, wherein the associated insertion value defines a correlation between the particular categorical variable and one or more insertion zone characteristics, and wherein the associated insertion value is determined using a machine learning module trained by insertion zone metadata determined by digital analysis of one or more instances of source video, from a pre-existing corpus of video, that each have at least one categorical variable in common with the new video;
• and
• determining, without performing analysis on image content of the new video, at least one predicted value of at least one insertion zone characteristic of the insertion zone metadata of the new video based at least in part on the associated insertion value.

A system for determining insertion zone metadata of a new video, the system comprising:

• one or more processors;
• and
• one or more memory modules, wherein the memory modules are configured to store computer executable instructions, which when executed on the one or more processors cause the system to:
• obtain, from data associated with the new video, categorical metadata relating to the new video, the categorical metadata comprising at least one categorical variable;
• obtain, for a particular categorical variable of the at least one categorical variable, an associated insertion value, wherein the associated insertion value defines a correlation between the particular categorical variable and one or more insertion zone characteristics, and wherein the associated insertion value is determined using a machine learning module trained by insertion zone metadata determined by digital analysis of one or more instances of source video, from a pre-existing corpus of video, that each have at least one categorical variable in common with the new video;
• and
• determine, without performing analysis on image content of the new video, at least one predicted value of at least one insertion zone characteristic of the insertion zone metadata of the new video based at least in part on the associated insertion value.