In practical application scenarios, the behavior of users watching movies is random and diverse, and also includes spatiotemporal features. Aiming at the fact that the complex ranking model cannot use a large amount of data for learning and updating in real time, especially the problem of insufficient training data for inactive users, this paper proposes a pre-training-based user embedding algorithm model. In the pre-training stage, the SINE model is used to dig out several intents with the highest user interest, improve the hit rate of user interest, and thus improve the accuracy of Inference. The follow-up test results show that the newly constructed recommendation model has better performance, and the evaluation index AUC is increased by 2.4% compared with the model without pre-training, which proves the effectiveness and feasibility of the new algorithm.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.