Mobile Location Profile Matching for Behave-Alike Audiences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately associate data sources with mobile devices, particularly in linking location data from mobile devices to individual or household profiles, which limits the effectiveness of targeted advertising and marketing strategies.
Innovation Solution
A system utilizing spatial and temporal analysis to build location profiles for mobile devices, associating them with addresses and purposes, and employing a supervised machine learning model to identify 'behave-alike' audiences based on demographic, behavioral, and spatial-temporal data, enabling more accurate targeting of advertising campaigns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing technologies are used to associate data sources with mobile devices, then the process is simple, but the accuracy of linking location data to individual or household profiles is poor
Solution Approach 1:
The system segments the data association process into distinct modules: location data collection, spatial analysis, temporal analysis, and profile association. Each module handles specific aspects of the complex task, making the overall system more manageable while improving accuracy through specialized processing at each stage.
Solution Approach 2:
The patent introduces intermediary data structures and processing layers that bridge location data and user profiles. These intermediaries include aggregated location histories, temporal patterns, and feature vectors that facilitate accurate matching without directly exposing raw data, thereby improving precision while managing complexity.
2Reliability
If location data is collected and analyzed to build profiles, then advertising targeting accuracy improves, but data processing complexity and computational requirements increase
Solution Approach 1:
The system performs preliminary processing of location data by aggregating and pre-analyzing spatial and temporal patterns before the actual advertising targeting. This preliminary action includes creating location histories, identifying patterns, and preparing feature vectors, which reduces the computational burden during the main targeting process while maintaining high reliability.
Solution Approach 2:
The patent transforms raw location data into standardized features and parameters that can be efficiently processed. By changing the representation of data from raw coordinates to processed features (such as location types, temporal patterns, and aggregated statistics), the system improves advertising effectiveness while reducing processing complexity through efficient data representation.
3Measurement precision
If comprehensive demographic and behavioral data is collected, then audience identification accuracy improves, but the amount of data to be processed and stored increases
Solution Approach 1:
The system extracts only the most relevant features and patterns from the comprehensive data set. By identifying and extracting key spatial and temporal features that are most predictive of audience behavior, the system maintains high identification accuracy while significantly reducing the volume of data that needs to be processed and stored, discarding redundant information.
Solution Approach 2:
The patent applies partial analysis by focusing computational resources on the most impactful data dimensions. Rather than processing all available data equally, the system identifies and processes only the critical features (such as location type, temporal patterns, and movement behavior) that contribute most to accurate audience identification, reducing overall data processing volume while maintaining precision.
Data Source
AI summary
Disclosed herein identifies audiences of mobile devices that behave a like a seed group of devices. That is, the behave alike group are those devices that move in similar patterns and visit the similar locations with a similar frequency as the devices of the seed group. Similarity is based on correlative similarity in having visited matching categories of location styles identified via mapping data (e.g., devices that visit national parks at a similar frequency). Correlative similarity is performed using a machine learning model trained via a follow the regularized leader proximal.