Query Anonymization for Efficient Online Concierge Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current online concierge systems face difficulties in analyzing queries with variable lengths due to the flexibility in the number of terms, which complicates the evaluation of processing efficiency and resource allocation.
Innovation Solution
The system extracts the syntactical structure of queries, anonymizes terms to account for the number of terms without using specific values, and generates representations using ranges to simplify grouping and analysis, allowing for more efficient evaluation and resource management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the system allows variable length queries with flexible numbers of terms, then user interaction is simplified, but query analysis and processing evaluation become more complex
Solution Approach 1:
The patent segments queries into two distinct components: syntactical structure (operators and field identifiers) and payload (terms with values). This segmentation allows the system to analyze and process queries independently by their structure rather than being overwhelmed by variable length variations, thus maintaining ease of use while reducing analysis complexity
Solution Approach 2:
The patent extracts the payload (terms with specific values) from the query and replaces it with a placeholder that indicates the number of terms present. This extraction removes the complexity of variable length terms from the analysis process while preserving the structural information needed for processing evaluation
2Adaptability or versatility
If the system processes queries with varying numbers of terms, then query flexibility is improved, but parallel processing and performance evaluation become difficult
Solution Approach 1:
The patent changes the representation parameter from the actual number of terms in the payload to a normalized indicator (such as logarithmic scale or bucketed ranges). This parameter transformation enables queries with different lengths to be grouped and processed in parallel efficiently, as the normalized parameters create uniform classification categories that facilitate parallel evaluation
3Measurement precision
If the system uses specific term values in queries, then query specificity is maintained, but anonymization and generalization for analysis become problematic
Solution Approach 1:
The patent extracts and removes specific term values from the query representation, retaining only the structural information (operators and field identifiers). The payload is replaced with a placeholder indicating the number of terms, which preserves the quantitative information needed for analysis while discarding the specific values that would complicate anonymization and generalization
Data Source
AI summary
When analyzing operation of an online system, such as an online concierge system, that receive queries from users and return results in response to those queries, queries received by the online system may include different numbers of terms. The variations in numbers of terms included in different queries makes it difficult to efficiently group queries for analysis, resulting in inefficient analysis of individual queries when analyzing different queries based on their structure. To more efficiently analyze different queries, the online system groups queries based on a number of terms in the query. In some embodiments, queries are grouped based on a logarithmic scale of the number of terms in the query, while in other embodiments queries are grouped based on a square root of the number of terms in the query.