Dynamic Card Ranking via Binary Relevance Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of electronic devices face unorganized information on web pages, making it difficult to find relevant content among multiple, unranked items, as existing systems lack personalized ranking based on user interests.
Innovation Solution
A system and method for ranking categories on a web page using binary outcomes, where a ranking server assigns relevance values to cards based on user interactions, determining the likelihood of relevance events, and adjusting the order of cards dynamically for each user based on their specific interests, without requiring domain-specific understanding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If information is presented in an unorganized manner on web pages, then all events are displayed simultaneously to users, but users cannot easily find information of interest
Solution Approach 1:
The patent changes the parameter of information organization from unorganized to organized by relevance ranking. The system dynamically adjusts the order of information presentation based on calculated relevance scores, transforming the static unorganized display into a dynamically optimized arrangement that prioritizes relevant content.
Solution Approach 2:
The patent replaces manual or simple chronological organization mechanisms with an automated relevance-based ranking system. The system uses computational algorithms to calculate relevance scores and automatically reorder information, substituting mechanical sorting with intelligent automated ranking.
2Adaptability or versatility
If a central team assigns relevance values to cards using various approaches, then different types of data can be presented, but the system complexity increases due to lack of universal card description
Solution Approach 1:
The patent creates a universal card structure that can accommodate multiple data types (video, sports, weather, etc.) through a standardized metadata format. This universal framework allows the same ranking system to handle diverse content types without requiring separate handling mechanisms for each card type.
Solution Approach 2:
The patent segments the card information into distinct metadata fields that can be independently evaluated. By dividing card data into structured components (title, description, category, etc.), the system can apply specific relevance rules to each segment while maintaining overall system simplicity.
3Measurement precision
If relevance values are assigned based on domain-specific understanding, then accurate relevance determination is achieved, but the central team cannot understand user wants in all knowledge domains
Solution Approach 1:
The patent enables the system to determine relevance automatically without requiring domain-specific expert understanding. The relevance ranking system serves itself by using objective metadata and user interaction data to calculate relevance scores, eliminating the need for central team members to have specialized knowledge in each domain.
Solution Approach 2:
The patent introduces metadata and automated algorithms as intermediaries between the content and the ranking system. These intermediaries translate domain-specific content into standardized relevance signals that the system can process objectively, bridging the gap between diverse knowledge domains and the universal ranking mechanism.
4Device complexity
If a broadcast order is provided to every user, then the system is simple to implement, but users do not find the web page more useful compared to personalized ordering
Solution Approach 1:
The patent transforms the static broadcast order into a dynamic personalized order. The system continuously adjusts the presentation order of cards based on individual user interactions and preferences, making the web page adaptive to each user rather than using a fixed universal order.
Solution Approach 2:
The patent implements feedback loops where user interactions (clicks, views, time spent) are continuously monitored and used to refine personalization. This feedback mechanism allows the system to learn from user behavior and improve relevance ranking over time, creating a progressively more useful experience.
Data Source
AI summary
A method includes accessing a number of cards from a database. The cards are ranked in the database based on a test conducted on a number of users. The cards are associated with one or more rule states. The one or more rule states provide binary outcomes of one or more rules. Each rule is identified using a code. The test is conducted by presenting different random sequences of the cards to different users and receiving inputs from the number of users. The method further includes receiving a request for a presentation area from a client device operated by a user. The presentation area is used for displaying the number of cards in an order, which is determined based on the test. The method includes providing the number of cards for display in the order within the presentation area on the client device of the user in response to the request.