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

Recommender systems play an important role in helping people find informationand make decisions in today's increasingly digitalized societies. However, thewide adoption of such machine learning applications also causes concerns interms of data privacy. These concerns are addressed by the recent "General DataProtection Regulation" (GDPR) in Europe, which requires companies to deletepersonal user data upon request when users enforce their "right to beforgotten". Many researchers argue that this deletion obligation does not onlyapply to the data stored in primary data stores such as relational databasesbut also requires an update of machine learning models whose training setincluded the personal data to delete. We explore this direction in the contextof a sequential recommendation task called Next Basket Recommendation (NBR),where the goal is to recommend a set of items based on a user's purchasehistory. We design efficient algorithms for incrementally and decrementallyupdating a state-of-the-art next basket recommendation model in response toadditions and deletions of user baskets and items. Furthermore, we discuss anefficient, data-parallel implementation of our method in the Spark StructuredStreaming system. We evaluate our implementation on a variety of real-worlddatasets, where we investigate the impact of our update techniques on severalranking metrics and measure the time to perform model updates. Our results showthat our method provides constant update time efficiency with respect to anadditional user basket in the incremental case, and linear efficiency in thedecremental case where we delete existing baskets. With modest computationalresources, we are able to update models with a latency of around0.2~milliseconds regardless of the history size in the incremental case, andless than one millisecond in the decremental case.

Efficiently Maintaining Next Basket Recommendations under Additions and Deletions of Baskets and Items