Browse Prior Art Database

Video Content and Social Comments based Advertisement Suggestion Disclosure Number: IPCOM000244416D
Publication Date: 2015-Dec-10
Document File: 3 page(s) / 32K

Publishing Venue

The Prior Art Database


Embedded advertising is basically product placement turned up a notch. Embedded advertising means the product part of the story a topic of discussion. Some embedded ads can be accepted by viewers and therefore the ads products got well known by people. However, some embedded ads always got negative comments. How to evaluate the appropriateness of the embedded ads given the specific movie/file plot from the audience point of view is always a challenge to most producers.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 86% of the total text.

Page 01 of 3

Video Content and Social Comments based Advertisement Suggestion

Main Idea:

Explore the comments of embedded ads from social media,and get the audience's judgements for the embedded ads.

Identify the implicit factors to embedded ads effect and build the embedded Ads scoring model

Guide the action of embedded ads for new movie/film/TV program.

Identify the features to evaluate the appropriateness embedded ads from audience's perspective.


Elements invloved in embedded ads
Product: feature, category etc
Role: personality,social position etc
Plot: people, location, topic etc

Features are defined in different kind of property space, and has no consistent standard to describe their similarity and relevance.

It is difficult to evaluate the suitablity and consistency of the embedded ads perspective between the heterogeneous factor of the embedded ads.


Page 02 of 3

Building embedded Ads scoring model

- Emedded ads evaluation from social comments

- Extract the social comments for embedeed ads

- Identify the elements from social comments, including, proudct, role

- Evalaute audience evaluation opinion for the embedded ads

- Semantic Relevance Space for embedded ads evalution

- Extract the product , role and plot from video segmentation,where to embed the ads

- Construct product , role and plot feature vector respectively

- Build up an uniform semantic space to depict product , role and plot with a consistent mode.

- Constuct the scoring model feature vector