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
Mr. DLib

Several new publications: Mr. DLib, Lessons Learned, Choice Overload, Bibliometrics (Mendeley Readership Statistics), Apache Lucene, CC-IDF, TF-IDuF

In the past few weeks, we published (or received acceptance notices for) a number of papers related to Mr. DLib, research-paper recommender systems, and recommendations-as-a-service. Many of them were written during our time at the NII or in collaboration with the NII. Here is the list of publications: Beel, Joeran, Bela Gipp, and Akiko Aizawa. “Mr. DLib: Recommendations-as-a-Service (RaaS) for Academia.” In Proceedings of the ACM/IEEE-CS Joint Conference on Digital Libraries (JCDL), 2017. Beel, Joeran. “Real-World Recommender Systems for Academia: The Gain and Pain in Developing, Operating, and Researching them.” In 5th International Workshop on Bibliometric-enhanced Information Retrieval (BIR) at the 39th European Conference on Information Retrieval (ECIR), 2017. [short version, official], [long version, arxiv] Beierle, Felix, Akiko Aizawa, and Joeran Beel. Read more…

By Joeran Beel, ago
Machine Learning

Mr. DLib v1.1.1 released: minor improvements

On 28th February, we released version 1.1.1 of Mr. DLib’s recommender system with some minor improvements and bug fixes: Improved 404 error handling for unknown document IDs Fix: The order of authors in the XML was not sorted properly Several internal changes (adjusted logging table; click time is not updated any more for second clicks etc;an automatic tool to add stereotype recommendations)

By Joeran Beel, ago
Mr. DLib

Farewell party for our visiting student Stefan Feyer

Stefan Feyer from the University of Konstanz had joined the Mr. DLib team for a six months internship. Yesterday, it was time to say goodbye to Stefan and we had a small farewell party at the National Institute of Informatics (NII) in Tokyo. We wish Stefan all the best for his future career and are grateful for all the work he did.

By Joeran Beel, ago
Machine Learning

Some numbers about Mr. DLib’s Recommendations-as-a-Service (RaaS)

Six months ago, we launched Mr. DLib’s recommendations-as-a-service for Academia. Time, to look back and provide some numbers: Since September 2016, Mr. DLib´s recommender system has delivered 60,836,800 recommendations to our partner Sowiport, and Sowiport’s users have clicked 91,545 of the recommendations. This equals on overall click-through rate (CTR) of 0.15%. The figure shows the number of delivered recommendations and CTR by month (2016-09-08 to 2017-02-11).  CTR is rather low and there is a notable variance among the months (e.g. 0.21% in September and 0.10% in December). The variance may be caused by different algorithms we are experimenting with. In addition, recommendations are also delivered when web spiders such as Google Bot are crawling our partner website Sowiport.de. In contrast, clicks are Read more…

By Joeran Beel, ago
Recommendations as-a-Service (RaaS)

Mr. DLib v1.1 released: JavaScript Client, 15 million CORE documents, new URL for recommendations-as-a-service via title search

We are proud to announce version 1.1 of Mr. DLib’s Recommender-System as-a-Service. The major new features are: A JavaScript Client to request recommendations from Mr. DLib. The JavaScript offers many advantages compared to a server-side processing of our recommendations. Among others, the main page will load faster while recommendations are requested in the background and a loading animation is shown. Using the JavaScript also means that the logging will be more reliable because web spiders are not logged any more. Our partner Sowiport uses the JavaScript already. We indexed 15 million documents from CORE and recommend them through our API. Another 5 million will follow soon. So far, recommendations could only be requested by specifying a particular document ID such as https://api-beta.mr-dlib.org/v1/documents/<ID>/related_documents/. Now, recommendations can Read more…

By Joeran Beel, ago
Recommendations as-a-Service (RaaS)

Enhanced re-ranking in our recommender system based on Mendeley’s readership statistics

Content-based filtering recommendations suffer from the problem that no human quality assessments are taken into account. This means a poorly written paper ppoor would be considered equally relevant for a given input paper pinput as high-quality paper pquality if pquality and ppoor contain the same words. We elevate for this problem by using Mendeley’s readership data for re-ranking Mr. DLib’s recommendations. This means, once we have a number of e.g. 20 documents that are related for a requested input paper, we re-rank the 20 documents based on the number of readers they have on Mendeley. The most read papers are then recommended. More details will follow.

By Joeran Beel, ago