Phone Number to User Mapping via Confidence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network service providers face challenges in building accurate demographics models at the individual user level due to the lack of information about which user is associated with a specific phone number within a household, leading to biased data and limited value for targeted marketing and service management.
Innovation Solution
A method and apparatus that assign unassigned phone numbers to specific users within a household based on confidence levels, using a combination of customer-provided data, third-party demographics, and usage measurements, with machine learning algorithms to predict the most likely user for each phone number.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If phone numbers are associated with a Billing Account Number at the household level, then multiple users can be served under a single account, but individual user identification is lost and demographics models become less accurate
Solution Approach 1:
The patent segments the household-level Billing Account Number into individual user-level phone number associations. By creating separate profiles for each user within the household account structure, the system enables both multi-user service under a single account and accurate individual user identification through machine learning-based phone number to user mapping.
2Productivity
If phone numbers are assigned to specific users, then targeted marketing and service management can be improved, but the complexity of managing user-phone number mappings increases
Solution Approach 1:
The system employs machine learning algorithms that automatically perform phone number to user identification without requiring manual configuration. The algorithms analyze usage patterns, location data, and device information to self-determine the most likely user for each phone number, reducing manual intervention while enabling targeted marketing capabilities.
3Measurement precision
If machine learning algorithms are used to predict user-phone number associations, then individual user identification accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The system applies machine learning algorithms selectively to phone numbers that cannot be directly associated with users through traditional methods. Rather than processing all phone numbers through complex algorithms, the system uses straightforward association methods where possible and applies machine learning only when needed, optimizing computational resource usage while maintaining identification accuracy.
Data Source
AI summary
A method and apparatus for assigning a mobile device that is unassigned to a user are disclosed. Mobile phone numbers are used as a surrogate for device. For example, the method implemented via a processor obtains the phone number that is unassigned from a list of phone numbers associated with a billing account number, assigns the phone number that is unassigned to the user, wherein the user is selected from among a plurality of users associated with the billing account number, determines a confidence level for the assigning of the phone number that is unassigned to the user, and performs an analysis to provide a service for the user to whom the phone number is assigned.