Abstract
The quality of user-contributed content at online platforms can vary. Manual validation of such content for quality and relevance is difficult and not scalable. No current techniques utilize known high quality content to rank user-contributed content. This disclosure describes techniques for scalable automatic validation of user-contributed content (UCC) provided to an online platform. A similarity function is learned based on merchant media and known high-quality user submitted media. The similarity function is used to score new UCC media and determine whether the new UCC media is relevant and of sufficient quality to include on the platform. The techniques improve the experience of using online platforms by ensuring that UCC is relevant to the entity with reference to which it is contributed and is of good quality. Further, the techniques can generate feedback on improving UCC contributions.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
S, Hari Bhaskar and Ketkar, Nikhil Shirish, "Automatic Validation of User-contributed Content Using Learned Similarity Function", Technical Disclosure Commons, (December 21, 2023)
https://www.tdcommons.org/dpubs_series/6530