Query Topic Category Transition Analysis for Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information retrieval systems fail to precisely analyze the transition between user query topic categories and relevant document topic categories, leading to inadequate user intention classification and query expansion, resulting in low user satisfaction and inability to detect attractively ranked documents.
Innovation Solution
A query and document topic category transition analysis system that classifies user queries and selected documents, generates a topic category transition map, and uses this analysis to expand queries by ranking representative keywords, thereby improving information retrieval results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional document similarity ranking is used, then information retrieval can be performed, but user preference and query category dependency are excluded from ranking factors
Solution Approach 1:
The system performs preliminary classification of queries and documents into topic categories before ranking. Topic category classifiers are applied in advance to both queries and documents, and transition probabilities are pre-computed, allowing the ranking to adapt to user preferences without adding complex real-time processing during query execution.
Solution Approach 2:
Topic categories serve as an intermediary layer between user queries and document ranking. Instead of directly comparing query and document similarity, the system uses topic category transition probabilities as a mediator to bridge user intent with relevant documents, enabling preference adaptation through a structured intermediate representation.
2Measurement precision
If topic category classification is applied to queries and documents, then user intention can be analyzed, but the transition between query topic category and document topic category cannot be precisely analyzed
Solution Approach 1:
The system uses feedback from user interactions (clicks, selections) to compute and refine topic category transition probabilities. When users select documents from retrieval results, this feedback is used to update the transition probability matrix, improving the precision of transition analysis over time while preserving user interest information through iterative learning.
3Measurement precision
If query expansion is performed without topic category transition analysis, then retrieval can be executed, but query expansion accuracy is insufficient
Solution Approach 1:
Topic category transition probabilities are computed in advance and stored in a matrix structure. During query expansion, the system performs a single matrix lookup based on the query's topic category, avoiding time-consuming real-time analysis while maintaining high expansion accuracy through pre-computed transition knowledge.
4Productivity
If conventional information retrieval ranking is used, then documents can be ranked, but attractively ranked documents cannot be detected
Solution Approach 1:
The system introduces topic category transition probability as a new ranking parameter that changes the traditional retrieval efficiency metric. By incorporating this parameter into the ranking function, the system can detect attractive documents that resonate with user interests while maintaining overall retrieval efficiency through efficient probability computations.
Data Source
AI summary
An information retrieval system and method, and more particularly, a query and document topic category transition analysis system and method in which a query topic category of a query input from a user as an information retrieval keyword and a document topic category of a document which a user regards as relevant and selects from information retrieval results are classified to analyze transition between the query topic category and the document topic category, and a query expansion-based information retrieval system and method using query and document topic category transition analysis in which a query input from a user is expanded using a topic category transition analysis result, and corresponding information or documents are retrieved using the expanded query are provided.