Automated Advertisement Target Inference Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current advertisement target selection methods are inefficient and costly, requiring manual data analysis by scientists, making it difficult for advertisers to quickly and affordably target suitable devices for their advertising needs.
Innovation Solution
An automated system that determines advertisement targets by obtaining usage history information from multiple devices, generating feature vectors, assigning labels based on advertisement requests, and using an inference model to identify suitable target devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual data analysis by data scientists is used to select advertisement targets, then the accuracy and precision of target selection is improved, but the time and cost increase significantly
Solution Approach 1:
The patent replaces manual data analysis by data scientists with an automated machine learning-based system. The system automatically extracts features from usage history information, generates feature vectors, determines labels based on advertisement requests, and uses an inference model to identify advertisement targets without human intervention, thereby reducing time and cost while maintaining accuracy
Solution Approach 2:
The patent transforms raw usage history information into structured feature vectors through automated feature extraction and transformation processes. By changing the data representation from raw logs to normalized feature vectors, the system enables efficient automated processing while preserving the essential information needed for accurate target selection
2Reliability
If manual data analysis by data scientists is used to select advertisement targets, then the quality of target selection is improved, but the cost increases significantly
Solution Approach 1:
The patent replaces expensive manual data analysis services with a cost-effective automated machine learning system. The system uses pre-trained models and automated feature extraction to achieve reliable target selection without requiring high salaries and extensive time investment from data scientists, thereby significantly reducing operational costs
Solution Approach 2:
The system performs self-service by automatically extracting features from usage history, generating feature vectors, determining labels, and identifying advertisement targets without requiring human expertise. This automation eliminates the need for costly manual analysis while maintaining selection quality through robust algorithmic processes
3Productivity
If automated system is implemented to determine advertisement targets, then the speed and efficiency of target selection is improved, but the complexity of the system increases
Solution Approach 1:
The patent divides the complex advertisement target selection process into distinct modular segments: feature extraction module, feature vector generation module, label determination module, and inference model module. Each segment handles a specific task independently, making the overall complex system manageable and easier to implement through standardized processing stages
4Measurement precision
If comprehensive analysis of usage history information is performed, then the accuracy of advertisement target selection is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent extracts only the essential features from comprehensive usage history information through automated feature extraction. By identifying and extracting only the most relevant features (such as user behavior patterns, device usage metrics, and interaction data) rather than processing all raw data, the system maintains high selection accuracy while significantly reducing computational resource consumption
Solution Approach 2:
The patent transforms comprehensive raw usage history data into condensed feature vectors through parameter transformation and dimensionality reduction. This conversion compresses large volumes of data into essential feature representations, enabling accurate target selection with reduced computational requirements and lower energy consumption
Data Source
AI summary
Provided is a method of determining an advertisement target according to an advertisement request, the method includes: obtaining usage history information from a plurality of devices, obtaining features of the plurality of devices, based on the usage history information, and generating feature vectors for the obtained features; determining labels for the plurality of devices, based on the advertisement request and the obtained features; generating an advertisement target inference model, based on the determined labels and the feature vectors; and determining at least one advertisement target device among the plurality of devices by applying the generated advertisement target inference model to the plurality of devices.