Neural Network Social Media Post Relevance Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current social media platforms lack the ability to effectively determine the relevance of user posts for predicting purchasing behavior, leading to inefficient targeting of advertisements and product information.
Innovation Solution
A method using neural networks to analyze social media posts, calculating relevance probabilities and purchasing probabilities, and transmitting targeted product-related information to users based on their behavior patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If social media platforms target users who mention products, then advertisement coverage is broad, but relevance and effectiveness of targeting deteriorates due to including non-purchasing users
Solution Approach 1:
The patent segments the user base by analyzing individual social media posts and assigning relevance probabilities to each post. Users are divided into target and non-target groups based on their post relevance scores, enabling precise identification of potential buyers without requiring complex analysis of entire user profiles or populations.
Solution Approach 2:
The patent introduces an intermediary classification system that uses trained classifiers to evaluate social media posts. This intermediary layer processes raw post data and transforms it into relevance probabilities, which then guide advertising decisions. This intermediary mechanism simplifies the overall system by providing a clear decision-making bridge between raw data and advertising actions.
2Measurement precision
If social media platforms analyze all user posts for purchasing intent, then identification accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-training classification models on historical social media data before deployment. Once trained, these classifiers can rapidly evaluate new posts in real-time without requiring complex analysis during actual advertising operations. This pre-processing of analytical capabilities significantly reduces processing time for ongoing post analysis.
Solution Approach 2:
The patent replaces manual or rule-based mechanical analysis of posts with automated neural network classifiers. These trained models automatically evaluate post relevance and predict purchasing intent without requiring time-consuming manual review or complex rule-based systems, thereby maintaining high precision while reducing analysis time.
3Productivity
If social media platforms use simple mention-based targeting, then system complexity remains low, but advertising effectiveness deteriorates due to including irrelevant users
Solution Approach 1:
The patent changes the key parameter for targeting from simple mention detection to relevance probability scoring. By training classifiers to evaluate multiple parameters within posts (context, sentiment, specificity) and combining them into a relevance score, the system achieves high advertising effectiveness. This parameter transformation enables precise targeting while keeping the system architecture relatively simple through automated classification.
Data Source
AI summary
Example implementations include a system and method of recognizing behavior of a user. In example implementations, a first post and at least one subsequent post indicative of a product and associated with a first social media account is obtained. A relevance probability is calculated for each of the obtained first post and the at least one subsequent post. The obtained first post and the at least one subsequent post are sequentially analyzed by a second neural network to determine output values relevant to probability of purchasing the product. A probability of purchasing the product is calculated based on the determined output values associated with each post and the calculated relevance probabilities. Product-related information is transmitted to the user associated with the obtained first post based on the determined probability of purchasing the product.