Search Suggestion Engine Using Entity Co-occurrence Knowledge Base
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines face challenges in providing accurate search results due to incomplete or misspelled search queries, leading to irrelevant results, as existing methods for enhancing queries are not sufficiently accurate or effective in suggesting related entities.
Innovation Solution
A method employing an entity co-occurrence knowledge base and fuzzy-score matching algorithms to extract entities from partial search queries, suggest completions, and present related entities, using a client/server architecture with an entity extraction module and fuzzy-score matching module to improve search query accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If search engines use traditional query processing methods, then system complexity remains low, but search result accuracy deteriorates due to incomplete or misspelled queries
Solution Approach 1:
The system performs preliminary entity extraction and fuzzy matching on user queries before full search processing. By pre-identifying entities and their types, and performing fuzzy matching against the knowledge base in advance, the system prepares corrected and enriched query versions beforehand, improving search accuracy without significantly increasing overall system complexity during runtime
Solution Approach 2:
The patent introduces an entity co-occurrence knowledge base as an intermediary between the user query and the search engine. This knowledge base stores pre-extracted entities, their types, and co-occurrence relationships, serving as a mediator that enables fuzzy matching and query correction without requiring complex real-time processing in the main search engine
2Adaptability or versatility
If search engines enhance queries using past query logs, then query enhancement is achieved, but suggestion accuracy deteriorates because log-based methods may lead users to results not of interest
Solution Approach 1:
Instead of deriving suggestions from past user queries (bottom-up approach), the system inverts the approach by building an entity co-occurrence knowledge base from structured data sources first, then using this pre-built knowledge to guide query enhancement. This top-down approach ensures suggestions are based on verified entity relationships rather than potentially noisy historical logs
Solution Approach 2:
The system performs preliminary entity extraction and co-occurrence analysis to build the knowledge base before it is needed for query enhancement. By pre-processing and structuring entity relationships in advance, the system has accurate reference data ready when users submit queries, enabling precise suggestions without relying on less accurate log-based methods
3Measurement precision
If search engines perform detailed entity extraction and fuzzy matching, then search suggestion accuracy improves, but processing time increases
Solution Approach 1:
The system performs entity extraction and co-occurrence analysis in advance to build the knowledge base, so that when queries are submitted, the heavy processing has already been completed. During query processing, the system only needs to perform fuzzy matching against the pre-built knowledge base, significantly reducing real-time processing time while maintaining high accuracy
Solution Approach 2:
The patent segments the search process into distinct phases: offline entity extraction and knowledge base construction, and online fuzzy matching and suggestion generation. By separating these tasks in time and function, the system can perform detailed analysis offline without impacting online response time, achieving both accuracy and speed
Data Source
AI summary
A method for generating search suggestions of related entities based on co-occurrence and/or fuzzy score matching is disclosed. The method may be employed in a search system that may include a client/server type architecture. The search system may include a user interface for a search engine in communication with one or more server devices over a network connection. The server device may include an entity extraction module, a fuzzy-score matching module, and an entity co-occurrence knowledge base database. In one embodiment, the search system may process a partial search query from a user and present search suggestions to complete the partial query. In another embodiment, the complete search query may be used as a new search query. The search system may process the new search query, run an entity extraction, find related entities from the entity co-occurrence knowledge base, and present said related entities in a drop down list.