Abstract: The technology involves a personalized recommender system that can be used with an e-commerce platform. It employs a contrastive learning based cross-domain recommendation approach. The approach balances the learning of user behaviors within each domain, as well as user behaviors across multiple domains. To achieve robust user representations and to improve knowledge transfer between the source and target domains, multi-task intra-domain contrastive regularizations may be employed along with multiple branches of sequential attentive encoders in a model for cross-domain sequential recommendation. Different data augmentation approaches can be used to generate augmented data for contrastive learning. For instance, different data augmentation methods may be combined with recommendation optimization in a multi-task learning paradigm. An optimized sequence representation may be fine-tuned in a next-value prediction task for recommendation in a target domain.

Abstract: The disclosure relates to a computer server system implementing a method to obtain a plurality of online articles for display on a webpage; obtain a candidate promoted content for each of the plurality of online articles; for each of the plurality of online article and the corresponding candidate promoted content pairs: determine a virality score of the online article indicating popularity of the online article among online users; determine a similarity score indicating similarity between the online article and the candidate promoted content; determine a qualification score based on the virality score and the similarity score; select a pair of target article and target promoted content from the plurality of article and candidate promoted content pairs based on the corresponding qualification scores; and display the target promoted content on the webpage.

Abstract: The disclosure relates to a computer server system implementing a method to obtain a plurality of online articles for display on a webpage; obtain a candidate promoted content for each of the plurality of online articles; for each of the plurality of online article and the corresponding candidate promoted content pairs: determine a virality score of the online article indicating popularity of the online article among online users; determine a similarity score indicating similarity between the online article and the candidate promoted content; determine a qualification score based on the virality score and the similarity score; select a pair of target article and target promoted content from the plurality of article and candidate promoted content pairs based on the corresponding qualification scores; and display the target promoted content on the webpage.