Utilizing a machine learning model to predict performance and generate improved digital design assets
US10789610B2 · kind B2 · utility
Assignee
Inventor
Key dates
| Filing date | Sep 8, 2017 |
| Grant date | Sep 29, 2020 |
| Priority date | — |
| Expiry date | Aug 16, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06Q30/0251
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Methods, systems, and computer readable storage media for improving predictive analytics for performance of digital design assets. In particular, one or more embodiments train a machine-learning model based on previously used digital design assets. One or more embodiments use the machine-learning model to analyze attributes of a user-generated digital design asset to generate an asset score that predicts the performance of the user-generated digital design asset for a target audience segment. One or more embodiments also use the machine-learning model to generate attribute scores for the attributes, and then generate the asset score based on the attribute scores. Additionally, one or more embodiments then provide the asset score to a user (e.g., a content creator) within an asset creation application to allow the user to improve the digital design asset for use in one or more digital content campaigns directed to the target audience segment.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.