Search Query Expansion via Knowledge Base and Statistical Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional electronic search systems using keyword searches are inefficient due to the dynamic nature of document corpora, where relationships between search terms become obsolete as documents and devices join or leave the network, leading to suboptimal search results and increased computing resource consumption.
Innovation Solution
A system utilizing knowledge base expansion with statistical models like Latest Semantic Indexing (LSI) to expand initial search terms, creating a comprehensive query by combining knowledge base relations and statistical correlations, allowing for real-time adaptation to changing document corpora and improved search efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If keyword search is used for electronic document search, then the search process is simple and fast, but the search results are incomplete with only 25% retrieval rate of relevant documents
Solution Approach 1:
The patent introduces an intermediary component (search expansion module) that sits between the user's keyword input and the document corpus. This module automatically expands keywords by querying a knowledge base for related terms, synonyms, and conceptually similar terms, thereby improving search completeness without requiring the user to manually specify multiple search terms.
Solution Approach 2:
The system performs preliminary action by pre-building a knowledge base containing relationships between terms, concepts, and documents before the actual search occurs. This knowledge base includes pre-computed relationships such as synonyms, hypernym-hyponym relationships, and co-occurrence patterns, allowing the search expansion to happen quickly during query processing rather than requiring real-time analysis of the entire corpus.
2Reliability
If knowledge base is updated frequently to reflect changing document corpora, then search relevance is maintained, but computing resource consumption increases
Solution Approach 1:
The system implements dynamic adaptation by continuously monitoring changes in the document corpus and automatically updating the knowledge base only when necessary. The search expansion module adapts to new terminology and relationships as documents are added or removed from the corpus, maintaining search relevance without requiring complete re-indexing or frequent full-updates of the knowledge base.
Solution Approach 2:
The patent employs parameter changes by adjusting the frequency and scope of knowledge base updates based on detected changes in the document corpus. Instead of rigid periodic updates, the system modifies its update parameters (such as update threshold, sampling rate, or expansion depth) dynamically based on corpus stability, thereby reducing unnecessary computing resource consumption while maintaining search quality.
3Productivity
If manual keyword selection is used for search, then the search process is simple, but only 25% of relevant documents are retrieved
Solution Approach 1:
The system implements self-service by automatically performing search term expansion without requiring user intervention. The search expansion module autonomously queries the knowledge base, retrieves related terms and concepts, and integrates them into the search query, eliminating the need for users to manually research and specify multiple relevant keywords while achieving comprehensive document retrieval.
Solution Approach 2:
The patent replaces the mechanical process of manual keyword selection with an automated information processing system. Instead of relying on user cognition and manual input of multiple keywords, the system uses computational methods (knowledge base querying, statistical analysis, and algorithmic expansion) to automatically generate comprehensive search terms, substituting human mental effort with automated intelligence.
Data Source
AI summary
Systems, devices, and methods include generating, for at least one search term of a query, a first expanded set of search terms based on at least one related term of the at least one search term obtained from a knowledge base, generating a second expanded set of search terms by applying a statistical model to the search terms of the first expanded set based on terms included in an electronic document corpus stored in an electronic data storage, and generating a third expanded set of search terms based on search terms in the first and second expanded sets. For each search term of the third expanded set, a term score is determined based on occurrence of the search term in the electronic document. The term scores are combined to create a document score for the electronic document.