Geo-Location Interest Inference via Weighted PoI Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately infer user interests from geo-location data due to noisy and uncertain factors such as coarse-grained or missing reports, irrelevant services, general population targeting, and distinguishing between current and remote locations of interest.
Innovation Solution
A system and method that retrieve point-of-interests (PoIs) from a repository based on geo-data items, generate weighted counts using pre-determined criteria, and aggregate scores to infer user interests, considering attributes like precision, accuracy, and proximity, to differentiate between relevant and noisy data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all geo data items are processed to infer user interests, then the quantity of analyzed data increases, but the accuracy of interest inference deteriorates due to noise from irrelevant services and casual user actions
Solution Approach 1:
The patent extracts and removes noise from the geo data by identifying and filtering out data from irrelevant services (weather services, news services, navigation services) and casual user actions (swiping, zooming). This extraction of harmful elements allows the system to process larger quantities of geo data while maintaining inference accuracy by excluding noisy portions.
Solution Approach 2:
The patent applies different quality standards to different portions of geo data based on their source and context. Relevant geo data from user-initiated location-based services are processed with high weight, while data from irrelevant services are filtered or weighted differently. This local differentiation of data quality allows simultaneous processing of diverse geo data volumes while maintaining overall inference accuracy.
2Area of stationary object
If geo-location data with coarse-grained reports is used, then the coverage of location data increases, but the precision of location identification deteriorates
Solution Approach 1:
The patent segments geo-location data into different precision levels and processes them accordingly. Coarse-grained reports are segmented and handled separately from fine-grained reports, allowing the system to maintain broad geographic coverage while identifying specific locations of interest with appropriate precision levels for each data type.
Solution Approach 2:
The patent changes the precision parameter of location identification based on the quality and granularity of the input geo data. When coarse-grained reports are received, the system adjusts its identification precision expectations and uses aggregation techniques to infer interests from broader location patterns, thereby maintaining coverage without requiring impossible precision.
3Measurement precision
If multiple attributes of geo data items are considered for weighting, then the accuracy of interest inference improves, but the complexity of the processing system increases
Solution Approach 1:
The patent segments the weighting process into distinct attributes (service type, user action type, location precision, temporal patterns) that can be evaluated independently. Each attribute contributes a separate weight factor, allowing the system to improve inference accuracy through multi-attribute consideration while maintaining manageable complexity through modular processing of each attribute type.
Data Source
AI summary
A method for inferring a user interest from geo data items associated with the user. The method includes retrieving point-of-interests (PoIs) from a PoI information repository based on the geo data items, generating a weighted count of the PoI for each geo data item that is weighted based on an attribute of the geo data item, and aggregating the weighted count across all geo data items to generate a score of the PoI, wherein the interest level of the user is inferred based at least on the score of the PoI.