To learn about our research interests, please read here https://www.scss.tcd.ie/joeran.beel/research/ and for a list of our publications see here https://www.scss.tcd.ie/joeran.beel/publications/

Datasets

Once a year, we publish Mr. DLib´s “Related-Article Recommendation Dataset” (RARD). RARD  is based on Mr. DLib´s recommender-system as-a-service from the digital library and reference management software domain. As such, it complements datasets from other domains such as books, movies, and music. The RARD II dataset encompasses 89m recommendations, covering an item-space of 24m unique items. RARD II provides a range of rich recommendation data, beyond conventional ratings. For example, in addition to the usual ratings matrices, RARD II includes the original recommendation logs, which provide a unique insight into many aspects of the algorithms that generated the recommendations. In this paper, we summarise the key features of this dataset release, describing how it was generated and discussing some of its unique features.

For more details, please visit http://data.mr-dlib.org and read our publications about RARD (II).

Living Lab

In cooperation with the reference management software JabRef, we provide a living lab for scholarly recommendations. This living lab allows you – as a research partner – to evaluate your novel recommendation algorithms in online evaluations with JabRef´s users. The workflow is as follows. When a user in JabRef selects the “related article” tab, JabRef sends a request to Mr. DLib´s API. Mr. DLib´s A/B engine randomly chooses whether to use Mr. DLib´s internal recommender system or your recommender system. If your system is chosen, Mr. DLib forwards JabRef´s request to your API, which returns a list of recommendations to Mr. DLib. Mr. DLib forwards the recommendations to JabRef. When a user clicks on a recommended document, a logging event it sent to Mr. DLib’s API and your system. Mr. DLib´s living lab is open for research partners whose recommender system is available through a REST API, which accepts a string as input (typically a source article’s title), returns a list of related-articles in JSON or XML, and includes URLs to web pages on which the recommended articles can be downloaded, preferably for free. Also, recommendations must be returned within less than 2 seconds.

If you want to participate in the living lab, please contact us.