Generalized Edit Distance for Query Expansion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face challenges in formulating effective search queries due to ambiguous wording or missing relevant terms, leading to suboptimal search results, and advertisers struggle to match user queries with relevant keywords for advertisements.
Innovation Solution
The method involves selecting query pairs from user sessions, determining co-occurrence values and transition costs between terms, and using these to generate query suggestions, expansions, or keyword matches that reflect semantic similarities and user intent, thereby improving search results and advertisement relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users enter simple or ambiguous search queries, then the ease of operation is improved, but the measurement precision of search intent deteriorates
Solution Approach 1:
The patent introduces an intermediary system (query analysis module with co-occurrence matrices) that mediates between the user's simple query input and the search engine's need for precise intent understanding. The system automatically analyzes query term co-occurrences and generates expanded queries without requiring users to manually refine their searches, thus maintaining ease of operation while improving measurement precision of search intent.
Solution Approach 2:
The system enables self-service by automatically performing query expansion and refinement using co-occurrence data from user sessions. The query expansion occurs autonomously based on learned patterns from user behavior, without requiring manual intervention from users to specify their intent more precisely, thereby maintaining ease of operation while improving search intent precision.
2Productivity
If query expansion is performed automatically, then the productivity of information retrieval is improved, but the device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing co-occurrence matrices from user session data before actual search queries are processed. The system learns and stores term co-occurrence patterns in advance, so that when a query arrives, the expansion can be performed quickly using pre-prepared data structures, improving productivity while managing complexity through offline preparation.
Solution Approach 2:
The system segments the query processing into distinct modules: query analysis, co-occurrence matrix lookup, and query expansion. This segmentation allows each component to be optimized independently and improves maintainability. The complexity is distributed across separate functional units rather than concentrated in a single complex system.
3Measurement precision
If transition costs are calculated based on co-occurrence values, then the measurement precision of term relationships is improved, but the use of energy increases
Solution Approach 1:
The system performs preliminary computation by pre-calculating co-occurrence values from user session data and storing them in matrices. This offline pre-computation shifts the energy-intensive operations away from real-time query processing, reducing the energy consumption during actual search operations while maintaining high measurement precision of term relationships through the pre-computed data.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining a generalized edit distance for queries. In one aspect, a method includes selecting query pairs of consecutive queries, each query pair being a first query and a second query consecutively submitted as separate queries, each first and second query including at least one term. For each query pair, the method includes selecting term pairs from the query pair, each term pair being a first term in the first query and a second term in the second query; and determining a co-occurrence value for each term pair. The method also includes determining transition costs based on the co-occurrence values for term pairs, each transition cost indicative of a cost of transitioning from a first term in a first query to a second term in a second query consecutive to the first query.