Abstract
Computer-implemented systems and methods for optimizing input selection for target systems such as automated product classifiers are provided. Such a system provides higher quality listings for buyers, more listing success for merchants and marketplaces, and increased revenue to marketing platforms. A model (e.g., analytical, heuristic, or machine-learned) is trained using an input such as images intended for product listings on a target system (e.g., marketing platform) and responses received from the target system. For future inputs (e.g., images), the model may generate an output such as a prediction, classification, or probability that an input will be accepted by the target system.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sanders, Douglas; Pang, Dave; Liao, Ying-hsiang; and Nguyen, Minh, "Optimized Input Selection for Target Systems", Technical Disclosure Commons, (November 04, 2022)
https://www.tdcommons.org/dpubs_series/5477