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/

Zendy Plus

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

Zendy Tools

Zendy Open

Conversational recommendation system (CRS) is able to obtain fine-grained anddynamic user preferences based on interactive dialogue. Previous CRS assumesthat the user has a clear target item. However, for many users who resort toCRS, they might not have a clear idea about what they really like.Specifically, the user may have a clear single preference for some attributetypes (e.g. color) of items, while for other attribute types, the user may havemultiple preferences or even no clear preferences, which leads to multipleacceptable attribute instances (e.g. black and red) of one attribute type.Therefore, the users could show their preferences over items under multiplecombinations of attribute instances rather than a single item with uniquecombination of all attribute instances. As a result, we first propose a morerealistic CRS learning setting, namely Multi-Interest Multi-roundConversational Recommendation, where users may have multiple interests inattribute instance combinations and accept multiple items with partiallyoverlapped combinations of attribute instances. To effectively cope with thenew CRS learning setting, in this paper, we propose a novel learning frameworknamely, Multi-Choice questions based Multi-Interest Policy Learning . In orderto obtain user preferences more efficiently, the agent generates multi-choicequestions rather than binary yes/no ones on specific attribute instance.Besides, we propose a union set strategy to select candidate items instead ofexisting intersection set strategy in order to overcome over-filtering itemsduring the conversation. Finally, we design a Multi-Interest Policy Learningmodule, which utilizes captured multiple interests of the user to decide nextaction, either asking attribute instances or recommending items. Extensiveexperimental results on four datasets verify the superiority of our method forthe proposed setting.

Multiple Choice Questions based Multi-Interest Policy Learning for Conversational Recommendation