Annotation of videos using aggregated user session data

A system for associating keywords to items of content based on search requests, comprising:

• one or more servers in communication with a data storage and with a plurality of second client devices, the one or more servers executing operations to
• receive, from a first client device, a first request for a first item of content;
• maintain, in a data storage, an association between the first item of content and keywords extracted from a plurality of Internet search requests transmitted by a plurality of second client devices within a predetermined temporal window of playback of the first item of content by the plurality of second client devices, wherein the keywords are extracted by
• identifying the plurality of second client devices on which the first item of content was presented;
• identifying the plurality of Internet search requests performed by the plurality of second client devices within the predetermined temporal window of playback of the first item of content by the plurality of second client devices;
• and
• selecting the keywords from the plurality of Internet search requests based on a frequency of the keywords included in the plurality of Internet search requests.

A non-transitory computer readable storage device storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

• receiving, from a first client device, a first request for a first item of content;
• identifying a plurality of second client devices on which the first item of content was presented;
• identifying a plurality of Internet search requests performed by the plurality of second client devices within a predetermined temporal window of playback of the first item of content by the plurality of second client devices;
• extracting keywords from the plurality of Internet search requests based on a frequency of the keywords included in the plurality of Internet search requests;
• and
• maintaining, in a data storage, an association between the first item of content and the keywords extracted from the plurality of Internet search requests.