Browse Prior Art Database

A system for adaptively identifying optimal insertion earmarks for media streams based on audience and peer group reactions

IP.com Disclosure Number: IPCOM000180257D
Original Publication Date: 2009-Mar-05
Included in the Prior Art Database: 2009-Mar-05
Document File: 2 page(s) / 20K

Publishing Venue

IBM

Abstract

Insertion of ads (or other materials) in media streams (e.g. audio, video) has been a popular method for targeting specific content (e.g. advertisements) to audiences for a long time now. Recently however, some companies (e.g. Google) have announced plans to support software that support the insertion of dynamic ads (e.g. replace one ad with another) based on various criteria (e.g. age of ad, targeting topics to specific audiences, etc.). There are some draw backs to these approaches however. For example, these insertion points are pre-defined locations in the media stream, and the inserted advertisements (and other types of content) can be seen as intrusive and annoying, ignored, or worse, result in a drop of audience viewer-ship (e.g. losing audience interest in a video because an inserted advertisement is too long).

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 2

A system for adaptively identifying optimal insertion earmarks for media streams based on audience and peer group reactions

Insertion of ads (or other materials) in media streams (e.g. audio, video) is a popular method for targeting specific content (e.g. advertisements) to audiences. Recently, however, some companies have announced plans to support software that supports the insertion of dynamic ads (e.g. replace one ad with another) based on various criteria (e.g. age of ad, targeting topics to specific audiences, etc.). There are some draw backs to these approaches however. For example, these insertion points are pre-defined locations in the media stream, and the inserted advertisements (and other types of content) can be seen as intrusive and annoying, ignored, or worse, result in a drop of audience viewer-ship (e.g. losing audience interest in a video because an inserted advertisement is too long).

This system leverages a Media Server that gauges audience reaction and social tolerance to identify points of interest (or displeasure) in a particular media stream playback, instead of the traditional method of relying on statically pre-defined locations. To gather this information, the system employs a series of extensible feedback mechanisms to identify insertion points. These include but are not limited to: 1) a deployed rating system where portions of the stream can be rated, 2) monitoring portions of the media stream that are replayed, or bookmarked, more times than others,
3) tracking where questions or comments are attached to media playback in certain time ranges, 4) monitoring playback abandonment (for example media player is closed), etc. These feedback loops are leveraged by the Media Server to determine optimal "hotspots" (or ranges) to optimally insert augmenting video to make it much less likely to be abandoned or ignored during playback.

In addition, advertising content is typically recorded in multiple lengths (60 second commercial vs. 15 second summary commercial) and the Media Server system could leverage this fact to determine the optimal length of video to b...