Contextual Advertising Database Using Semantic Metadata Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital advertising systems face challenges in balancing effective targeting with user privacy concerns, particularly due to stringent regulations like GDPR and e-Privacy, which restrict the use of personal data for behavioral targeting, leading to resource-intensive and often ineffective data collection and processing.
Innovation Solution
A system utilizing a classification module to determine primary weights for data streams, a recognition module to identify explicit and implicit concepts, and a storage module to save these concepts in a database, allowing for targeted advertising without personal tracking data, using weighted semantic metadata to match content with advertising campaigns and determine optimal placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If personal data collection and processing is used for behavioral targeting, then advertising targeting accuracy is improved, but user privacy protection deteriorates and regulatory compliance becomes difficult
Solution Approach 1:
The patent extracts and removes personal identifiable information from the data processing pipeline. Instead of collecting and processing personal data for targeting, the system processes only anonymous aggregated data that cannot be traced back to individual users, thereby eliminating privacy violations while maintaining targeting capabilities through contextual analysis
Solution Approach 2:
The patent introduces an intermediary layer of contextual analysis that mediates between advertising goals and user privacy. Rather than directly linking ads to individual user behavior, the system uses contextual metadata about content and anonymous aggregated user preferences as an intermediary, enabling targeting without personal data exposure
2Adaptability or versatility
If extensive personal data collection is performed, then behavioral targeting capability is improved, but resource consumption increases
Solution Approach 1:
The patent extracts only the essential contextual features needed for targeting while discarding unnecessary personal data. By processing only anonymous aggregated data and contextual metadata rather than complete personal profiles, the system reduces data volume and processing energy requirements while maintaining behavioral targeting capability
Solution Approach 2:
The patent applies partial action by processing only the subset of data necessary for contextual targeting rather than complete personal data sets. The system processes anonymous aggregated preferences and contextual content metadata, which is sufficient for effective targeting without the excessive resource consumption of processing full personal data sets
3Productivity
If personal data is stored and processed, then advertising effectiveness is improved, but regulatory compliance risk increases
Solution Approach 1:
The patent extracts and removes personal identifiable information from the advertising system. By using only anonymous aggregated data for targeting decisions, the system maintains advertising effectiveness through contextual matching while eliminating GDPR and other privacy regulation compliance risks associated with personal data storage and processing
Solution Approach 2:
The patent introduces anonymous aggregated user profiles and contextual metadata as intermediaries between advertising campaigns and users. This intermediary layer enables effective targeted advertising while ensuring regulatory compliance, as the intermediary data cannot be traced back to individual users and thus does not trigger privacy regulation requirements
Data Source
AI summary
A system for creating and/or maintaining a database is disclosed. In one example, the system includes one or more processors; a classification module configured to determine primary weights for primary data streams, each primary weight referring to a correlation between one of the primary data streams and one segment category of several predefined segment categories; a recognition module configured to identify explicit concepts and implicit concepts in the primary data streams, and to determine first secondary weights characterizing embeddings of the identified concepts; an expansion module configured to determine for the identified concepts respective related concepts; and a storage module configured to save the identified concepts.


