6th Recommender Meetup Amsterdam

Author: Andreas Lommatzsch

The Recommender Meetup Amsterdam (www.recommenders.nl) is meetup focusing on innovative recommender algorithms and AI methods. The meetup is organized by the TU Delft and hosted by different companies in the Netherlands. The 6th RecSys:NL Meetup was held on September 28th at ORTEC Amsterdam.

There were three talks: First, Alexandros Karatzoglou (Telefonica Research, Barcelona, Spain) gave a talk about Deep Learning Methods and the use of Deep Leaning approaches in Recommender Systems. In his talk, he discussed the state of the art of Deep Learning methods and explained current trends and application scenarios in the Deep Learning methods are used.

The second talk was given by Andreas Lommatzsch (TU Berlin, Berlin, Germany). In his talk "NewsREEL challenge: Online evaluation of recommender systems" he explained the CLEF NewsREEL challenge and the challenges of online evaluation methods. The CLEF NewsREEL challenge allows researchers to evaluate algorithms for recommending online news articles. The talk explained the NewsREEL scenario in detail. The specific challenges have been discussed such as the continuous changes in the set of relevant items, the influence of context parameters, and the variance in the user behavior. In addition, the talk presented different algorithmic approaches as well as the experiences gained organizing the challenge. The slides are available here.

Daniel Kohlsdorff (XING, Hamburg, Germany) explained the challenges of providing high-quality job recommendations. He presented selected components of the XING’s recommender system explaining content based recommendations with latent spaces, and outlier filtering. In addition, Daniel presented the results of the RecSys challenge 2016 (co-organized by XING) and gave an outlook on the RecSys challenge 2017.

After the talks there has been a lively discussion accompanied by pizza and beer.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>