Associative Intent Query Discovery from Search Logs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines lack the ability to effectively suggest alternative queries and provide diverse advertisements based on the associative intent of user search queries, as they primarily rely on matching key words without considering relationships between queries.
Innovation Solution
A system and method for discovering associative intent queries by mining user sessions from query logs, generating query pairs, removing similar pairs, and using statistical and semantic tests to identify associated query pairs, which can be used to offer search suggestions and diverse advertisements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If search engines rely on keyword matching to retrieve information, then the search process is simple and fast, but the ability to suggest alternative queries and provide diverse advertisements is limited
Solution Approach 1:
The system performs preliminary mining of user sessions and generation of query pairs before actual search queries are processed. By pre-analyzing query logs and identifying associative relationships in advance, the search engine can prepare suggestion candidates that will be quickly retrieved when needed, reducing real-time processing complexity while maintaining high adaptability.
Solution Approach 2:
The search processing is segmented into distinct phases: user session mining, query pair generation, similarity removal, and associative intent identification. This segmentation allows each phase to be optimized independently, with the complex associative intent discovery performed offline during session mining, while online search processing remains simple and fast.
2Measurement precision
If search engines analyze user sessions and query relationships, then search suggestion accuracy improves, but processing time and computational resources increase
Solution Approach 1:
User session mining and query pair generation are performed as preliminary actions during offline processing. By completing these time-consuming tasks in advance, the system achieves high measurement precision for search suggestions without incurring the processing time cost during actual user queries, effectively separating computation intensity from response time.
Solution Approach 2:
The system creates a simplified representation of user sessions and query relationships as data structures that can be quickly queried during online search operations. By copying and pre-processing the complex query log data into optimized data formats, the system maintains high suggestion accuracy while reducing real-time computational requirements.
3Reliability
If similar query pairs are removed from the dataset, then the quality of associative intent queries improves, but the quantity of available query pairs decreases
Solution Approach 1:
The system extracts and removes similar query pairs from the dataset through a dedicated filtering process. By taking out redundant and similar query pairs, the system improves the reliability and quality of remaining associative intent queries, ensuring that only distinct and meaningful query associations are retained for advertisement targeting.
Data Source
AI summary
A system is described for discovering associative intent queries based on search web logs. The system may mine one or more user sessions comprising data from search engine query logs and generate query pairs based on the data. The system may use statistics and morphology to identify relationships among the query pairs. From these relationships, the system may distinguish the associative intent query pairs from the similar and unrelated intent query pairs.