Search Result Ranking Using User Interaction Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines face challenges in providing high-quality search results due to the presence of low-quality documents in large databases, vulnerability to spamming techniques, and failure to effectively utilize user interaction time as a relevance indicator.
Innovation Solution
A computer-implemented method that determines the length of time users spend on search results, using this temporal information to rank and refine search results for subsequent searches, incorporating it as a factor in determining relevance alongside other metrics like page rank and content similarity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If search engines use traditional ranking methods (vector space model, citation counting), then search results can be generated efficiently, but the quality and relevance of results deteriorate due to spamming and low-quality documents
Solution Approach 1:
The patent implements feedback by measuring user interaction time with search results and using this information to adjust and improve future search rankings. The system continuously monitors how long users spend on search result pages and feeds this data back into the ranking algorithm to refine result quality over time
Solution Approach 2:
The system uses user behavior data (time spent on pages) as self-generated feedback to automatically improve its own search quality without external intervention. The search engine serves itself by leveraging its own collected user interaction data to refine its ranking algorithms
2Quantity of substance
If users are presented with hundreds of search results to improve coverage, then more potential relevant information is available, but the selectivity and quality of results deteriorate due to irrelevant documents camouflaging relevant ones
Solution Approach 1:
The patent changes the ranking parameter from traditional metrics (keyword matching, citation count) to user interaction time. By measuring and weighting the duration users spend on search result pages, the system transforms how results are prioritized, placing highly relevant results at the top where users are most likely to engage
Data Source
AI summary
A system and method for producing more relevant search results. When a user selects a search result from a search result listing, the amount of time that the user spends interacting with the item associated with the search result is tracked. Such information regarding interaction time is compiled and is used as a factor in assessing relevance of items in future searches.