3D Model Activity Ranking for Geographic Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network-based search services face challenges in effectively ranking geographic locations based on user interest, as they lack a comprehensive method to quantify and prioritize locations based on user engagement metrics such as 3D model uploads, clicks, points of interest, and embedded views.
Innovation Solution
A computer-implemented method and system that determine user interest values for geographic locations by analyzing metrics like the number of 3D models, user requests, points of interest, and embedded views associated with those locations, and orders search results accordingly, using a weighted sum to prioritize locations with higher user engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional search ranking algorithms are used, then search results can be generated quickly, but the relevance and user interest in the results deteriorates
Solution Approach 1:
The ranking system is segmented into multiple independent scoring components: 3D model activity score, embedded view score, point of interest score, and traditional relevance score. Each component calculates a specific aspect of location quality independently, then they are combined through weighted summation. This segmentation allows the system to incorporate complex user engagement metrics without overwhelming computational complexity.
Solution Approach 2:
The system performs preliminary calculations by pre-computing user interest values for locations based on 3D model activity, embedded views, and points of interest before the actual search query is processed. These pre-computed values are stored and reused across multiple searches, reducing real-time computational burden while maintaining high relevance ranking accuracy.
2Measurement precision
If comprehensive user engagement metrics are analyzed, then search result accuracy improves, but processing time increases
Solution Approach 1:
User interest values incorporating 3D model activity, embedded view counts, and point of interest data are pre-computed and stored for each location. When a search query arrives, the system retrieves these pre-computed values rather than calculating them from scratch, significantly reducing processing time while maintaining comprehensive analysis of user engagement metrics.
Solution Approach 2:
The system selectively applies the comprehensive ranking algorithm only to locations that pass initial filtering based on traditional relevance criteria. For locations with low traditional relevance, the system uses simpler ranking methods, thereby reducing overall processing time while maintaining high accuracy for the most relevant results where comprehensive analysis matters most.
Data Source
AI summary
Aspects of the invention relate generally to ranking geographic locations based on perceived user interest. More specifically, a database of three-dimensional models of buildings or other architectural features may be used to determine the level of user interest in a particular location and accordingly rank, for example, geographic locations or web or map search results with local intent. For example, various signals such as the number of models created by users for a particular location, the number of clicks or requests for the models of the particular location, the number of POIs contained within or associated with the models associated with the particular geographic location, number of categories associated with a model associated with the geographic location, number of embedded views or views of the models associated with the particular location on other web sites, and the age of the models associated with the particular geographic location.