Wireless Hotspot POI Matching Using User Scanning Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for matching wireless hotspots with Points of Interest (POIs) are inefficient and costly, with limited data collection speed and accuracy, and often rely on manual collection or user feedback, which can be inaccurate.
Innovation Solution
A method and apparatus that use user-scanned hotspot position information to acquire candidate POIs and rank them based on access characteristic information, such as usage patterns, to determine the matching POI without manual data collection or user feedback, utilizing machine learning models like LambdaMART for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dedicated employees are hired to collect matching information between wireless hotspots and POIs, then data collection accuracy can be ensured, but collection efficiency is low and costs are high
Solution Approach 1:
The system enables automatic self-service data collection by utilizing user-generated hotspot scanning data and automatically matching it with POI information through algorithms, eliminating the need for manual collection by dedicated employees while maintaining high accuracy
Solution Approach 2:
The system implements feedback mechanisms where user scanning behavior data is continuously collected and used to refine and update the matching algorithms, improving both accuracy and efficiency through iterative optimization based on real-world usage patterns
2Reliability
If dedicated employees are hired to collect matching information, then data quality can be controlled, but collection costs increase significantly
Solution Approach 1:
Instead of employing dedicated personnel for data collection, the system creates and utilizes copies of user scanning data automatically generated during normal app usage, transforming user behavior into valuable training data without additional collection costs
Solution Approach 2:
The system performs automatic data collection and processing through algorithms that run autonomously, eliminating labor costs associated with dedicated employees while maintaining data quality through automated validation and filtering mechanisms
3Measurement precision
If POI matching is based on SSID of wireless hotspot, then matching accuracy is improved, but coverage is reduced because not all POIs have directly related SSID names
Solution Approach 1:
The system transitions from relying solely on SSID name matching (one dimension) to incorporating multiple dimensions including user scanning behavior patterns, temporal information, spatial relationships, and POI category data, enabling both high accuracy and broad coverage simultaneously
Solution Approach 2:
The system dynamically adjusts matching parameters and weights based on different scenarios and POI types, allowing flexible adaptation to various situations while maintaining overall system accuracy and expanding coverage to include POIs without directly related SSID names
Data Source
AI summary
In some embodiments, a method includes: obtaining position information of a wireless hotspot where the user locates, based on hotspot scanning information of wireless hotspots already scanned by the user; acquiring one or more candidate Points of Interest (POIs) close to the wireless hotspot based on the position information; based on characteristic information corresponding to the wireless hotspot and respective candidate POIs, ranking all candidate POIs to determine an POI matching the wireless hotspot, wherein the characteristic information comprises access characteristic information. In some embodiments, the following advantages may be realized: the POI matching the wireless hotspot is obtained based on relevant data that the user scans the wireless hotspot to predict the POI actually accessed by the user without a procedure of manually collecting the data or user feedback, thereby improving the efficiency.