Concept Map Translation for Cookie-Free Network Location Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for understanding user activities online require significant time and effort to build cookies profiles, leading to processing overhead in bid requests, and there is a need for a more efficient way to understand users based on the content they view.
Innovation Solution
A concept map is used to translate identifiers in bid requests into target profiles without relying on cookies, by mapping URL profiles to predefined nodes, enabling real-time processing without tracking historical user data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cookies are used to track user activities and build user profiles, then user understanding accuracy is improved, but processing time and computational effort increase significantly
Solution Approach 1:
The system pre-generates concept maps and URL profiles before bid requests arrive. Concept maps are built by crawling and analyzing web content in advance, and URL profiles are created by mapping URLs to these pre-built concept maps. This preliminary preparation eliminates the need for real-time cookie processing, allowing bid requests to be handled quickly by simply matching URL profiles against ad target profiles using the pre-computed concept maps.
Solution Approach 2:
The invention extracts and removes the cookie dependency from the bid request processing system. Instead of relying on cookies to carry user profile information, the system extracts user context directly from the URL being accessed and maps it to pre-computed concept maps. This extraction of cookie dependency eliminates the need for storing and processing large volumes of cookie data in real-time.
2Adaptability or versatility
If cookies are used to track user activities, then user profiling capability is improved, but device complexity and storage requirements increase
Solution Approach 1:
The system creates simplified copies of user context information through URL profiles that reference pre-computed concept maps. Instead of storing complete user activity histories in cookies, the system creates lightweight URL profiles that contain only the essential mapping between URLs and concept map nodes. These profiles are much smaller and simpler than traditional cookie-based user profiles, reducing storage requirements and system complexity while maintaining profiling capability.
Solution Approach 2:
The concept map serves as an intermediary structure that mediates between URLs and user profiles. Rather than directly storing and querying large volumes of user activity data in cookies, the system uses concept maps as an intermediate layer that pre-organizes web content semantics. This intermediary structure simplifies the matching process between URL profiles and ad target profiles, reducing the computational complexity of user profiling.
3Productivity
If real-time user profiling is performed without cookies, then processing speed is improved, but profiling accuracy may deteriorate
Solution Approach 1:
The system performs preliminary analysis of web content to build comprehensive concept maps before bid requests arrive. During this offline phase, the system crawls web pages, extracts semantic information, and constructs detailed concept maps that capture the relationships between different web content elements. This preliminary action ensures that when bid requests arrive in real-time, the system can quickly match URL profiles against these pre-computed concept maps without sacrificing profiling accuracy, achieving both speed and precision.
Data Source
AI summary
An example operation may include one or more of storing target profiles associated with respective targeted content, receiving a request comprising a content identifier of content on a network being accessed on a user device, identifying a content profile corresponding to the content identifier, mapping the concepts included in the identified content profile to a target profile among the stored target profiles via a concept map, and transmitting targeted content associated with the mapped target profile.