Recommender Systems

Our website ranks #1 for ‘recommender systems ireland’ and ‘recommender systems dublin’ searches on Google

We started working at Trinity College Dublin 1.5 years ago and launched our new website only 2 months ago. Yet, Google ranks our website #1 for the search queries ‘recommender systems ireland‘ and ‘recommender systems dublin‘ and, not surprisingly, for the variations ‘ireland recommender systems‘ and ‘dublin recommender systems‘. Of course, this is not to mean that we, the School of Computer Science and Statistics at Trinity College Dublin or the ADAPT Centre are the undisputed authorities in the field of recommender systems in Dublin or Ireland. There are several notable researchers and institutions more including Prof. Barry Smyth, Dr Derek Bridge and the Insight Centre. However, this good Google ranking is a flattering approval of our work in the field of recommender systems. For more details on our work please Read more…

By Joeran Beel, ago
Conferences

New publication: Position Bias in Recommender Systems for Digital Libraries

We recently had a paper accepted to iConference. We used click data from Mr DLib, our recommender-as-a-service, to see if users of recommender systems in digital libraries are affected by position bias. We found that users are affected by position bias on average. Some users do seem to examine recommendations more carefully than other users. As far as we know, this is the only investigation of position bias in digital libraries, and in recommender systems for digital libraries. We have made the data available. It contains 10 million recommendations delivered by us to users, with 12,543 clicks on recommendations logged. We hope that other researchers will find this useful. For example, perhaps the click behavior of a digital library user could Read more…

By Andrew Collins, ago
Jobs / Career

We Are Hiring: 1 Software/Machine-Learning Engineer & 1 Software Architect / Product Owner for a Recommender-System Business Start-up

UPDATE: We will soon advertise another position for this start-up. Please come back in a few days. The School of Computer Science and Statistics of Trinity College Dublin and the ADAPT Centre received funding to hire 2 employees for 2 years* to spin-out a business start-up in the field of recommender-systems as-a-service and machine learning in Dublin. The two positions are to be filled with one machine-learning engineer and one software architect/product manager, whereas both employees are expected to work together very closely. They will be responsible for developing a recommender-system as-a-service that uses a unique technology, based on the research of Prof Dr Joeran Beel who will be the project lead (read here for a brief outline of the Read more…

By Joeran Beel, ago
Recommender Systems

Seminar by Prof Dietmar Jannach: “Recommender Systems – Beyond Matrix Completion”

We are delighted to announce a seminar by Prof Dr Dietmar Jannach on Recommender Systems at Trinity College Dublin. Dietmar Jannach is a well-known researcher in the field of recommender systems and author of the book “Recommender Systems: An Introduction“. The seminar is open to all staff, students, and visitors in Dublin who are interested in recommender systems. Title: Recommender Systems – Beyond Matrix Completion Abstract: Automated recommendations have become a common part of our daily online user experience. Significant advances were made in recent years in terms of algorithmic approaches to compute recommendations for users. The main task in such an algorithm-focused setting is to predict through machine learning approaches how relevant a certain item will be for an Read more…

By Joeran Beel, ago
Mr. DLib

Mr. DLib Recommendations-as-a-Service v1.3: “Word Embeddings” and Many Minor Improvements and Bug Fixes

We released version 1.3 of Mr. DLib´s Recommender-System as-a-Service. The new major feature is “word embeddings” based recommendations. We are excited to see how the new recommendations will perform with our partners. In addition, we fixed many small bugs, and added some minor improvements.  A complete overview can be found in JIRA.

By Joeran Beel, ago
Conferences

Report from the 11th ACM Conference on Recommender Systems

We just returned from the 11th ACM Conference on Recommender Systems in Como, Italy. It was an amazing conference, with lots of interesting presentation relating to recommender systems. One of the hot topics at the Recommender Systems Conference was Deep Learning, though, frankly, deep learning did not always seem to deliver promising results for recommender systems. Here are a few photos.

By Joeran Beel, ago
Mr. DLib

Mr. DLib v1.2.1: Improved keyphrase recommendations and Apache Lucene query handling

The new version of our recommender system completes 104 issues and significantly improves the recommendations. The most notable improvements are: We improved the keyphrase extraction process in the recommender system, i.e. keyphrases are not stored differently in Lucene. We expect better recommendation effectiveness and are currently running an A/B test. More robust path encoding for search queries (special characters in a URL caused errors) Lucene’s eDismax function is A/B tested (together with Lucene’s standard query parser) Improved queries for CORE recommender system (their system needs queries to be of a certain length; Mr. DLib now just multiplies the queries until they are at least 50 characters) Abstracts and keywords in the XML response of Mr. DLib are enclosed in <![CDATA[ HTML Snippet is improved Read more…

By Joeran Beel, ago
Mr. DLib

RARD: The Related-Article Recommendation Dataset

We are proud to announce the release of ‘RARD’, the related-article recommendation dataset from the digital library Sowiport and the recommendation-as-a-service provider Mr. DLib. The dataset contains information about 57.4 million recommendations that were displayed to the users of Sowiport. Information includes details on which recommendation approaches were used (e.g. content-based filtering, stereotype, most popular), what types of features were used in content based filtering (simple terms vs. keyphrases), where the features were extracted from (title or abstract), and the time when recommendations were delivered and clicked. In addition, the dataset contains an implicit item-item rating matrix that was created based on the recommendation click logs. RARD enables researchers to train machine learning algorithms for research-paper recommendations, perform offline evaluations, and Read more…

By Joeran Beel, ago
Mr. DLib

Mr. DLib 1.2 released: JabRef recommendations completed; CORE recommendation API connected

There are two major news coming along with the new version of Mr. DLib’s Recommendation API. JabRef finally uses Mr. DLib for it’s recommender system We have announced this already a while ago, but now, finally, Mr. DLib’s recommendations are available in one of the most popular open-source reference managers, i.e. JabRef. Currently, Mr. DLib enables JabRef users to retrieve a list of related-article recommendations, given a currently selected entry in the reference list (see screenshot). In the long run, we aim for creating personalized recommendations, too. Mr. DLib is not the only provider of recommendations-as-a-service in Academia. Another provider is the CORE project, with whom we partnered now. CORE is offering an API similar to the one we offer. We Read more…

By Joeran Beel, ago