Search Query Clustering for Local Intent Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital content systems lack the ability to effectively infer and respond to the location intent behind search queries, leading to inefficient content distribution and user experience.
Innovation Solution
The system clusters search queries based on matching features and determines a local intent threshold, assigning intent flags to query clusters, allowing for the modification of content items with local features for high local intent queries and omitting them for low local intent queries, thereby enhancing user experience and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If local content features are added to all search query responses, then user experience for location-based queries is improved, but bandwidth usage and system resource consumption increase
Solution Approach 1:
The patent applies local quality by differentiating content delivery based on query characteristics. High local intent queries receive enhanced content with local features (maps, directions, business information), while low local intent queries receive standard content without these features. This selective approach ensures that bandwidth is consumed only when locally-relevant content is actually needed, rather than universally adding local features to all search results.
2Measurement precision
If the system infers local intent from query clusters, then content distribution accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the search query space into clusters based on shared characteristics and local intent patterns. By grouping queries with similar features together, the system can infer local intent at the cluster level rather than analyzing each individual query in isolation. This segmentation approach improves content distribution accuracy while managing system complexity through batch processing and pattern recognition.
Solution Approach 2:
The system performs preliminary analysis by pre-processing search queries into clusters and pre-determining which clusters exhibit high local intent characteristics. This preliminary action allows the system to have local intent determination ready before actual content delivery occurs, improving response accuracy without adding complexity during the critical content delivery moment.
3Reliability
If local intent threshold is set low, then more queries receive local content features improving relevance, but unnecessary content is distributed increasing waste
Solution Approach 1:
The patent employs parameter changes by adjusting the local intent threshold to optimize the balance between content relevance and waste reduction. By carefully calibrating this threshold parameter, the system ensures that local content features are added only when the inferred local intent is sufficiently strong, thereby improving content relevance for users who need it while avoiding unnecessary content distribution that would waste bandwidth and resources.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining cluster data, determining a portion of the search queries within the given query cluster that trigger a local search feature, assigning to the intent flag a value indicating that the given cluster is high local intent cluster when the portion of the search queries within the given query cluster that trigger the local search feature meets the local intent threshold, assigning to the intent flag a value indicating that the given cluster is a low local intent cluster when the portion of the search queries within the given query cluster that trigger the local search feature fails to meet the local intent threshold, and modifying a content item including adding a local content feature to the content item when the search query is determined to be included in the high local intent cluster.