Interest Graph Generation for Privacy-Safe Consumer Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for targeted consumer engagement over-rely on demographic and geographic data, which are limited by privacy concerns and do not directly predict consumption behavior, while failing to harness the utility of consumer interest information.
Innovation Solution
Generating and using an interest graph that represents subjects' affinities towards topics, free of personally identifiable information, by analyzing self-stated and self-revealing interests from various interactions, and incorporating demographic data in aggregated form to create a visual representation of relatedness between interests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If businesses rely on demographic and geographic data for targeted advertising, then it is easy to obtain consumer information through loyalty programs and surveys, but the effectiveness of engagement is limited and privacy concerns arise
Solution Approach 1:
The patent transitions from using demographic parameters (age, gender, location) to using interest-based parameters (topics, affinities, consumption preferences) to represent consumers. This parameter change enables more effective targeting by aligning advertising with actual consumer interests rather than statistical demographics, thereby resolving the contradiction between ease of data collection and engagement effectiveness
Solution Approach 2:
The patent introduces an interest graph as an intermediary data structure that connects consumer interests to content. This interest graph serves as a mediator between raw consumer data and targeted advertising delivery, enabling effective engagement while maintaining privacy by working with aggregated interest patterns rather than individual demographic profiles
2Loss of information
If businesses collect detailed demographic and geographic data, then they can create targeted advertisements, but consumer privacy is compromised and data protection becomes difficult
Solution Approach 1:
The patent extracts only the essential interest information from consumer data while leaving out personally identifiable demographic and geographic information. By taking out only the necessary affinity data toward topics and products, the system maintains advertising effectiveness while protecting consumer privacy, thus resolving the contradiction between information retention and privacy protection
3Ease of operation
If businesses use conventional demographic data for targeting, then data collection is straightforward through surveys, but the data does not directly predict consumption behavior
Solution Approach 1:
The patent performs preliminary analysis of consumer interactions with content, products, or services to extract interest information before using it for advertising purposes. By pre-processing consumer behavior data to identify affinities and preferences, the system achieves accurate consumption behavior prediction while maintaining operational simplicity, resolving the contradiction between ease of collection and prediction accuracy
Data Source
AI summary
Methods and apparatuses are provided for generating, updating, and using an interest graph. A plurality of interests representing a plurality of subjects' affinities towards a plurality of topics may be obtained. A processing device may generate an interest graph based on the obtained interests. The generated interest graph may include: (i) at least two nodes, each of the at least two nodes representing an interest of the obtained interests, wherein the obtained interests are free of any personally identifiable information associated with the plurality of subjects and (ii) at least one link connecting a first node to a second node of the at least two nodes, the at least one link representing a relatedness of the interest represented by the first node to the interest represented by the second node.