Consumption Probability Metrics for Media Asset Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to accurately predict the number of unique users who will consume specific media assets, particularly those with specific preferences, and struggle to estimate exposure to multiple media assets in campaigns.
Innovation Solution
A viewership forecasting application generates consumption probability metrics by combining aggregated and disaggregated viewership data, calculating weights, and modifying probabilities to predict user consumption and exposure to media assets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If aggregated viewership data is used to predict consumption, then the prediction covers total audience size, but it cannot identify cross-sections of users with specific preferences
Solution Approach 1:
The patent segments the total audience into individual user profiles with specific characteristics and preferences. Each user is assigned a probability of consumption based on their profile attributes, enabling precise targeting of cross-sections (e.g., users who like trucks) while maintaining the overall aggregate prediction accuracy.
2Measurement precision
If traditional monitoring methods are used, then channel consumption can be identified, but unique user exposure to multiple media assets cannot be accurately predicted
Solution Approach 1:
The system uses aggregated forecast feedback to adjust individual user probabilities. The modification factor, derived from comparing aggregated predictions with actual consumption patterns, is applied to refine the probability estimates for each user, improving the accuracy of unique user exposure predictions across multiple media assets.
3Measurement precision
If disaggregated user-level probabilities are calculated, then cross-section prediction improves, but computational complexity increases
Solution Approach 1:
The system pre-calculates and stores user profiles with their characteristics and preferences before the forecasting process. This preliminary organization of data allows for efficient retrieval and probability calculation during media asset consumption prediction, reducing real-time computational complexity while maintaining high precision.
Data Source
AI summary
A consumption probability metric may be generated for a media asset. An aggregated forecast predicting user consumption of a media asset is received. A plurality of probabilities, each corresponding to a user of a plurality of users, is received, each indicating how likely a respective user is to consume the media asset. A weight for the plurality of users is calculated representing a ratio of the total number of users to a number of users in the plurality of users. A disaggregated forecast predicting user consumption of a media asset is determined based on the weight for the plurality of users and the plurality of probabilities. A modification factor is computed based on the aggregated forecast and the disaggregated forecast. A metric is generated that includes a plurality of user identifiers associated with the plurality of users and a plurality of modified probabilities each modified by the modification factor.