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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

An accurate prediction of watch time has been of vital importance to enhanceuser engagement in video recommender systems. To achieve this, there are fourproperties that a watch time prediction framework should satisfy: first,despite its continuous value, watch time is also an ordinal variable and therelative ordering between its values reflects the differences in userpreferences. Therefore the ordinal relations should be reflected in watch timepredictions. Second, the conditional dependence between the video-watchingbehaviors should be captured in the model. For instance, one has to watch halfof the video before he/she finishes watching the whole video. Third, modelingwatch time with a point estimation ignores the fact that models might giveresults with high uncertainty and this could cause bad cases in recommendersystems. Therefore the framework should be aware of prediction uncertainty.Forth, the real-life recommender systems suffer from severe bias amplificationsthus an estimation without bias amplification is expected. Therefore we proposeTPM for watch time prediction. Specifically, the ordinal ranks of watch timeare introduced into TPM and the problem is decomposed into a series ofconditional dependent classification tasks which are organized into a treestructure. The expectation of watch time can be generated by traversing thetree and the variance of watch time predictions is explicitly introduced intothe objective function as a measurement for uncertainty. Moreover, weillustrate that backdoor adjustment can be seamlessly incorporated into TPM,which alleviates bias amplifications. Extensive offline evaluations have beenconducted in public datasets and TPM have been deployed in a real-world videoapp Kuaishou with over 300 million DAUs. The results indicate that TPMoutperforms state-of-the-art approaches and indeed improves video consumptionsignificantly.

Tree based Progressive Regression Model for Watch-Time Prediction in Short-video Recommendation