Object Rank Authority Graph Search for Keyword Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital search technologies face challenges in providing thorough and relevant search results without including irrelevant information, particularly in keyword-based searches within databases, where traditional methods fail to differentiate between global importance and keyword-specific relevance.
Innovation Solution
The Object Rank system employs authority-based search techniques by modeling databases as labeled graphs, using authority transfer rates and semantic relationships to generate keyword-specific rankings, and incorporates information retrieval (IR) style relevance ranking to prioritize results based on relevance scores and structural information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional keyword-based search methods are used, then search speed is improved, but search relevance deteriorates because they cannot differentiate between global importance and keyword-specific relevance
Solution Approach 1:
The patent segments the search ranking problem into two distinct components: global importance ranking (using authority-based methods like PageRank) and keyword-specific relevance ranking (using IR-style scoring). This segmentation allows each component to be optimized independently and then combined to produce the final ranking, resolving the contradiction between speed and relevance.
Solution Approach 2:
The patent implements a dynamic ranking system that adapts to different query types and contexts. The hybrid ranking mechanism dynamically adjusts the weighting between global importance and keyword-specific relevance based on the query characteristics, allowing the system to maintain high relevance across diverse search scenarios while preserving computational efficiency.
2Loss of information
If comprehensive search results are provided to ensure thoroughness, then search completeness is improved, but information overload increases due to irrelevant information
Solution Approach 1:
The patent applies local quality by differentiating the treatment of different results based on their specific characteristics. High-authority documents containing keywords receive enhanced ranking, while other documents are ranked according to their relevance. This localized quality adjustment ensures comprehensive coverage while filtering out irrelevant information, preventing information overload.
Solution Approach 2:
The hybrid ranking system incorporates feedback mechanisms where the authority scores and relevance scores mutually inform the final ranking. The system continuously refines rankings based on the interaction between global authority metrics and keyword-specific relevance metrics, ensuring that comprehensive results are prioritized by quality rather than quantity, thus reducing information overload.
3Measurement precision
If authority-based search is used to improve relevance, then computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing authority scores (such as PageRank values) for all documents in the database. These pre-computed authority scores are stored and reused across multiple queries, eliminating the need to recalculate them for each search. This preliminary computation significantly reduces the computational complexity of subsequent searches while maintaining high relevance through authority-based ranking.
Solution Approach 2:
The patent merges two ranking approaches (authority-based ranking and IR-style relevance ranking) into a unified hybrid ranking system. By combining these methods with appropriate weighting, the system achieves high search relevance without the full computational burden of purely authority-based methods, as the IR component provides efficient keyword matching that reduces the reliance on computationally intensive authority calculations for every query.
Data Source
AI summary
This application describes methods for searching digital information such as digital documents (e.g., web pages) and computer databases, and specific search techniques such as authority ranking and information retrieval (IR) relevance ranking in keyword searches. In some implementations, the technique includes analyzing digital information viewed as a labeled graph, including nodes and edges, based on a flow of authority among the nodes along the edges, the flow of authority being derived at least in part from different authority transfer rates assigned to the edges based on edge type schema information. In some implementations, the system includes an object rank module configured to generate multiple initial rankings corresponding to multiple query keywords, each of the multiple initial rankings indicating authority of nodes in a graph with respect to each respective query keyword individually; and a query module configured to combine the multiple initial rankings in response to a query.