English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

This paper investigates Cross-Domain Sequential Recommendation (CDSR), apromising method that uses information from multiple domains (more than three)to generate accurate and diverse recommendations, and takes into account thesequential nature of user interactions. The effectiveness of these systemsoften depends on the complex interplay among the multiple domains. In thisdynamic landscape, the problem of negative transfer arises, where heterogeneousknowledge between dissimilar domains leads to performance degradation due todifferences in user preferences across these domains. As a remedy, we propose anew CDSR framework that addresses the problem of negative transfer by assessingthe extent of negative transfer from one domain to another and adaptivelyassigning low weight values to the corresponding prediction losses. To thisend, the amount of negative transfer is estimated by measuring the marginalcontribution of each domain to model performance based on a cooperative gametheory. In addition, a hierarchical contrastive learning approach thatincorporates information from the sequence of coarse-level categories into thatof fine-level categories (e.g., item level) when implementing contrastivelearning was developed to mitigate negative transfer. Despite the potentiallylow relevance between domains at the fine-level, there may be higher relevanceat the category level due to its generalised and broader preferences. We showthat our model is superior to prior works in terms of model performance on tworeal-world datasets across ten different domains.

Cracking the Code of Negative Transfer: A Cooperative Game Theoretic Approach for Cross-Domain Sequential Recommendation