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

In many recommendation applications such as news recommendation, the itemsthat can be rec- ommended come and go at a very fast pace. This is a challengefor recommender systems (RS) to face this setting. Online learning algorithmsseem to be the most straight forward solution. The contextual bandit frameworkwas introduced for that very purpose. In general the evaluation of a RS is acritical issue. Live evaluation is of- ten avoided due to the potential loss ofrevenue, hence the need for offline evaluation methods. Two options areavailable. Model based meth- ods are biased by nature and are thus difficult totrust when used alone. Data driven methods are therefore what we consider here.Evaluat- ing online learning algorithms with past data is not simple but somemethods exist in the litera- ture. Nonetheless their accuracy is not satisfac-tory mainly due to their mechanism of data re- jection that only allow theexploitation of a small fraction of the data. We precisely address this issuein this paper. After highlighting the limita- tions of the previous methods, wepresent a new method, based on bootstrapping techniques. This new method comeswith two important improve- ments: it is much more accurate and it provides ameasure of quality of its estimation. The latter is a highly desirable propertyin order to minimize the risks entailed by putting online a RS for the firsttime. We provide both theoretical and ex- perimental proofs of its superioritycompared to state-of-the-art methods, as well as an analysis of the convergenceof the measure of quality.

Improving offline evaluation of contextual bandit algorithms via bootstrapping techniques