Search Recall Set Sizing Based on Query Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines in online concierge systems are inefficient, wasting computational power and introducing unwanted latency due to ineffective management of search result sets.
Innovation Solution
An online concierge system employs a search module that dynamically adjusts the recall set size based on query entropy, using historical data and machine learning to optimize the number of search results, balancing latency and search result quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional search engines execute searches with fixed large result set sizes, then search result completeness is improved, but computational power consumption increases and latency increases
Solution Approach 1:
The patent applies dynamics by making the recall set size adjustable rather than fixed. The search module dynamically determines the recall set size based on query entropy calculations, allowing the system to adapt the result set size to the specific characteristics of each search query, thereby optimizing the balance between result completeness and computational efficiency
Solution Approach 2:
The patent changes the parameter of recall set size from a static fixed value to a dynamic value that varies based on query entropy. By calculating query entropy and using it to determine the appropriate recall set size, the system optimizes computational resource usage while maintaining search result quality
2Quantity of substance
If conventional search engines execute searches with fixed large result set sizes, then search result completeness is improved, but search latency increases
Solution Approach 1:
The system dynamically adjusts the recall set size based on query entropy to reduce search latency. By calculating entropy and determining an optimized recall set size, the system avoids the unnecessary processing time associated with fixed large result sets while ensuring sufficient search results are returned
Solution Approach 2:
The patent changes the recall set size parameter from fixed to dynamic, using query entropy as the determining factor. This parameter change enables the system to optimize search execution time by selecting appropriate result set sizes tailored to each query's characteristics
3Reliability
If the search module returns more search results, then search result quality is improved, but computational resources are wasted
Solution Approach 1:
The patent optimizes the recall set size parameter based on query entropy to balance search result quality with computational efficiency. By calculating the entropy of each query and using it to determine the appropriate result set size, the system avoids wasting computational resources on queries that do not require large result sets
Data Source
AI summary
A search module for an online concierge system executes searches in response to a search query with respect to item databases of retailers. The search module dynamically configures a recall set size that controls a number of search results returned for a search query based in part on a query entropy representing an estimated breadth of the search term. The query entropy may be determined relative to a diversity of items in a retailer's database. The recall set size may be configured relative to the query entropy in a manner that manages a tradeoff between latency of search execution and search result quality.