Video Ad Targeting via Web Page Content Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video hosting systems face challenges in targeting advertisements effectively when uploaded videos lack sufficient metadata, as direct content analysis is inefficient and may not accurately select ads for videos embedded on web pages with minimal content.
Innovation Solution
A computer-implemented method and system that identifies content locations embedding a video, aggregates content from multiple web pages, and uses derived keywords to select and transmit targeted advertisements for presentation with the video, even if the web page has minimal content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct content analysis is used to select advertisements for videos, then ad targeting accuracy may be improved, but computational efficiency and processing time deteriorate
Solution Approach 1:
The patent extracts and utilizes metadata and textual content from web pages that embed the video, separating this information from the video content itself. By analyzing only the surrounding text and metadata rather than the entire video, the system achieves adequate ad targeting without the computational burden of full video analysis.
Solution Approach 2:
The system performs preliminary analysis of web page content and metadata before video processing. By pre-extracting relevant textual information and keywords from the embedding context, the system prepares ad targeting data in advance, avoiding the need for complex real-time video analysis.
2Measurement precision
If video content is analyzed to select advertisements, then ad relevance to video content is improved, but computational resources and storage requirements increase
Solution Approach 1:
The patent extracts only the necessary textual elements (metadata, page content, keywords) from the web embedding environment, discarding the need to analyze the actual video content. This extraction approach maintains ad relevance while dramatically reducing computational resource requirements.
Solution Approach 2:
The system uses inexpensive, easily obtainable textual data from web pages as a substitute for expensive video analysis. The metadata and surrounding text serve as disposable proxies that provide sufficient information for ad selection without requiring heavy computational investment.
3Productivity
If advertisements are targeted based on web page content, then ad selection speed is improved, but ad accuracy for videos with minimal page content deteriorates
Solution Approach 1:
The patent merges content from multiple web pages that embed the same video to create a composite profile. By combining textual information, metadata, and keywords across several sources, the system compensates for the limitations of any single page with minimal content, maintaining both speed and accuracy.
Solution Approach 2:
Instead of relying solely on the content depth of individual web pages, the system transitions to a multi-dimensional approach by aggregating data across multiple pages. This dimensional expansion from single-page to multi-page analysis provides sufficient signal for accurate ad targeting even when individual pages are sparse.
4Measurement precision
If content from multiple web pages is aggregated to select advertisements, then ad targeting accuracy for videos with minimal content is improved, but system complexity increases
Solution Approach 1:
The patent segments the ad selection process into distinct modules: identifying embedding pages, extracting content and metadata, aggregating information across pages, and selecting ads based on the composite profile. This segmentation manages system complexity by breaking down the complex aggregation task into manageable, independent steps.
Data Source
AI summary
The subject matter of this specification can be embodied in, among other things, a method that includes identifying web pages that embed a selected video, aggregating content derived from the identified web pages, and transmitting one or more advertisements (ads) that are targeted to the aggregated content for presentation with the selected video embedded on the identified web pages.