Influencer Recommendation Model for Cosmetic Sales Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting social media content creators to advertise cosmetic products are often ineffective, as they rely on popularity metrics that may not accurately reflect the product's potential sales impact, and can be influenced by inactive or fake accounts and demographic mismatches.
Innovation Solution
A machine learning model is trained on historical account and product data to predict the utilization rate of a cosmetic product on a particular social media account, generating recommendations for partnering with content creators based on predicted sales performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If companies choose content creators based on popularity metrics (number of followers), then the selection process is simple and quick, but the accuracy of predicting actual product sales impact is low
Solution Approach 1:
The patent replaces manual evaluation methods with an automated machine learning system that processes historical data to predict utilization rates. The system substitutes human judgment and simple follower-count metrics with computational algorithms that analyze multiple parameters including account characteristics, product attributes, and historical performance data to generate accurate sales impact predictions.
2Adaptability or versatility
If manual analysis is used to evaluate content creators, then the analysis can be customized and flexible, but it requires significant time and resources
Solution Approach 1:
The system enables automated self-service analysis where the machine learning model independently processes historical data, evaluates content creators, and generates predictions without requiring manual intervention. The system serves itself by automatically training on historical utilization rate data and applying the trained model to evaluate new content creator-product pairings, eliminating the need for time-consuming manual analysis while maintaining flexibility through customizable input parameters.
3Ease of operation
If simple follower count metrics are used to select content creators, then the selection process is easy to implement, but it is influenced by inactive or fake accounts and demographic mismatches
Solution Approach 1:
The patent transforms the selection process by changing from a single parameter (follower count) to multiple parameters including account characteristics, product attributes, and historical utilization rates. The machine learning model processes these changed parameters to generate a comprehensive prediction of sales impact, making the selection process reliable while remaining operationally simple through automated model application.
Data Source
AI summary
A computer-implemented method for generating a recommendation for displaying a new cosmetic product on a particular account comprising obtaining training data including historical account parameters associated with a plurality of historical accounts, historical cosmetic product parameters associated with one or more historical cosmetic products displayed on the respective historical accounts of the plurality of historical accounts, and historical utilization rate data associated with the one or more cosmetic products; training, based on the training data, a machine learning model to predict utilization rates of cosmetic products, resulting in a trained machine learning model; applying the trained machine learning model to parameters associated with a particular account and parameters associated with a new cosmetic product to predict a utilization rate for the new cosmetic product if displayed on the particular account; and generating a recommendation based on the predicted data for displaying the new cosmetic product on the particular account.