Object Popularity Ranking via Graph Propagation Factors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engine technologies struggle to accurately rank objects within web pages based on their popularity, as they treat entire web pages as the smallest unit for retrieval and ranking, failing to distinguish the popularity of objects referenced within those pages.
Innovation Solution
A ranking system that calculates object popularity based on relationships between objects using a graph representation with nodes and links, applying a popularity propagation factor to each relationship and employing a page ranking algorithm to determine object-level popularity, allowing for intra-type and inter-type relationship analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If web pages are treated as the smallest unit for retrieval and ranking, then the ranking process is simplified, but the ability to accurately reflect object popularity within web pages is lost
Solution Approach 1:
The patent segments web pages into individual objects (products, people, papers, organizations, etc.) and ranks each object separately based on its popularity signals. This allows the system to measure object popularity independently from page popularity, resolving the contradiction by maintaining simplicity at the object level while achieving precision in popularity measurement.
Solution Approach 2:
The patent introduces a new dimension of analysis by distinguishing between page-level popularity and object-level popularity. Instead of ranking only at the page level, the system creates an additional dimension of object-level ranking that captures the popularity of specific entities within pages, thereby achieving measurement precision without oversimplifying the process.
2Measurement precision
If object-level popularity ranking is implemented, then accurate object popularity assessment is achieved, but the system complexity increases
Solution Approach 1:
The system segments the ranking task into distinct components: extracting objects from web pages, identifying popularity signals for each object, and calculating popularity scores. This segmentation manages complexity by breaking down the complex object-level ranking task into manageable, independent steps that can be processed separately.
Solution Approach 2:
The patent introduces popularity signals as an intermediary mechanism that bridges web page content and object popularity assessment. These signals (such as links, mentions, and other indicators) serve as mediators that capture object popularity without requiring direct analysis of all web page content, thereby reducing system complexity while maintaining measurement precision.
3Ease of operation
If entire web pages are ranked, then the ranking process is straightforward, but objects within unpopular pages cannot be identified as popular
Solution Approach 1:
The patent extracts objects from web pages and separates their popularity assessment from the page-level ranking. By taking out objects and evaluating them independently based on their own popularity signals, the system prevents loss of object popularity information that would occur if only page-level ranking were performed.
Solution Approach 2:
The system segments the evaluation process into page-level and object-level components. This segmentation allows the simple page-level ranking to coexist with more detailed object-level popularity assessment, ensuring that popular objects are not lost within unpopular pages while maintaining the simplicity of the overall ranking framework.
Data Source
AI summary
A method and system for ranking objects of different object types based on their popularity is provided. A ranking system calculates the popularity of objects based on relationships between the objects. A relationship indicates how one object is related to another object. Thus, objects of one object type may have one or more relationships with objects of another object type. One goal of the ranking system is to rank the objects of the different object types based on their popularity. The objects and their relationships can be represented using a graph with nodes representing objects and links representing relationships between objects. The ranking system assigns a popularity propagation factor to each relationship to represent its contribution to the popularity of objects of that type.