Latest News
Machine Learning

Machine Learning Meetup at Accenture (Competitions, Overfitting and Personality Traits) — 26th March

My colleague Declan McKibben from the ADAPT Research Centre organizes the monthly Machine Learning Meetup in Dublin, which takes place on the last Monday every month. This month, the machine learning meetup is hosted by Dogpatch and there will be plenty of exciting presentations about machine learning: Mick Cooney (Senior Analytics Consultant, Barnett Waddingham LLP) speaks about the Eldorado ML competition, which is similar to Kaggle competitions but focusing more on real-world aspects of machine learning. Atul Nautiyal (Post Doctoral Researcher, ADAPT Centre) speaks about one of the most common problems in machine learning, i.e. overfitting, and how to deal with overfitting using ‘dropout’. Scott Nowson (Consumer Analytics and AI Manager at Accenture) presents how to use machine learning to recognise the personality traits of individuals. The next machine learning meetup takes place on 26th March at The Dock, Accenture 7 Hanover Quay, Grand Canal Dock, Dublin. Drinks are served from 18:00, Read more…


Guest Lecture / Workshop by Thorsten Schaeff, Stripe

As part of our e-Business II lecture, we invited Thorsten Schaeff to give a guest lecture, more precisely, a workshop. Thorsten is a partner engineer at Stripe, which is a payment platform, competing with PayPal. During the 90 minute lecture, Thorsten will give a short introduction to Stripe, the history, their culture, and plans for Dublin as an engineering hub. We will then look at the “old way” of making payments on the Web, including the PCI standard, and then dive into a hands-on workshop focusing on building frictionless checkout experiences with Stripe Elements, their UI components for collecting credit card details and enabling wallet payments via Apple and Google Pay. If time allows we can then look at adding local payment methods like AliPay, and WeChat Pay to reach customers in Asia. The lecture will be in a workshop style, i.e. you will need your laptop, and you should register Read more…


Our new website is live!

Today, we launched our new website It provides lots of information about our research, publications, projects, and teaching relating to recommender systems, machine learning and more. The new website also combines the blog posts of our project websites Mr. DLib and Docear.

Jobs / Career

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

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 recommendations-as-a-service and machine learning. The two positions are to be filled with one software 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. The new project will be based on the existing prototype Mr DLib. However, the current recommender system prototype is supposed to be re-designed, re-implemented, and extended with various features, mostly relating to enhanced machine-learning algorithms. The goal for the two-year period is to develop a market-ready recommender system prototype, acquire around 10 more pilot partners, and spin-out a start-up company Read more…

Recommender System

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. Dietmar Jannach is a well-known researcher in the field of recommender systems and author of the book “Recommender Systems: An Introduction“. 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 individual user. Usually, the underlying problem is abstracted to one of completing a sparse user-item rating matrix. In this talk, I will introduce the basic concepts of recommenders and then discuss why being able to make accurate rating predictions not necessarily leads to the most useful recommendations in practice. We will then Read more…

Jobs & Internships

We welcome two DAAD interns in recommender systems and machine learning

As part of the DAAD RISE Worldwide program, we were awarded two funded internship positions for two undergraduate students both being from Germany. Gordian (University of Munich / LMU) and Martin (Universty of Göttingen) will spend around three months with us over the summer — Gordian at the National Institute of Informatics in Tokyo, and Martin at the ADAPT Centre and School of Computer Science at the Trinity College Dublin. They will be conducting a research project as part of Mr. DLib in the fields of recommender systems, machine learning and natural language processing.


Call for Papers: ACM Recommender Systems Conference 2018 in Vancouver

The ACM Conference on Recommender Systems (RecSys) is the premier conference to present new research results, systems, and techniques including machine learning and natural language processing relating to recommender systems.  This year, the 12th ACM Conference on Recommender Systems takes place in Vancouver, and the call for papers has just been published. We are proud to serve on the program committee and hope to receive many interesting submissions this year.


Update for Docear’s “Google Scholar Parser” Library to Fetch Metadata for PDF files

Google Scholar recently changed its layout, and as a consequence, Docear couldn’t fetch metadata anymore from Google Scholar for PDF files. Fortunately, one of our users (“Silberzwiebel”) adjusted Docear’s Google Scholar Parser, and now everything works as usual. However, we have not yet integrated the function into the main version of Docear. This means, even if you have just downloaded Docear, you need to manually update the Google Scholar Parser, if you want to fetch metadata for your PDF files. To update the Google Scholar parser, do the following Close Docear if it’s currently running Download the updated library docear-metadata-lib-0.0.1.jar. Replace the existing file “docear-metadata-lib-0.0.1.jar” with the new one. You will find the file in C:\Program Files (x86)\Docear\plugins\org.docear.plugin.bibtex\lib\ (Windows 10) or a similar directory, depending on your operating system. Start Docear, and fetch metadata 🙂


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.

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 (better layout for recommendations in JabRef), i.e. spaces were added, and “NULL” elements are not shown anymore For both queries and Lucene indexes, only lowercase is used (previously, we used Read more…

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 do research on data from Mr. DLib’s recommender system, without implementing a recommender system themselves. In the field of scientific recommender systems, our dataset is unique. To the best of Read more…

Contact Details

To send a message, use the contact form or send an email to


Trinity College Dublin, Department of Computer Science and Statistics, O'Reilly Institute,  College Green, Dublin 2, Ireland


Office G.15


Detailed directions are avilable in this PDF

Message Us