Document Embeddings vs. Keyphrases vs. Terms: An Online Evaluation in Digital Library Recommender Systems

05/27/2019
by   Andrew Collins, et al.
0

Many recommendation algorithms are available to digital library recommender system operators. The effectiveness of algorithms is largely unreported by way of online evaluation. We compare a standard term-based recommendation approach to two promising approaches for related-article recommendation in digital libraries: document embeddings, and keyphrases. We evaluate the consistency of their performance across multiple scenarios. Through our recommender-as-a-service Mr. DLib, we delivered 33.5M recommendations to users of Sowiport and Jabref over the course of 19 months, from March 2017 to October 2018. The effectiveness of the algorithms differs significantly between Sowiport and Jabref (Wilcoxon rank-sum test; p < 0.05). There is a  400 difference in effectiveness between the best and worst algorithm in both scenarios separately. The best performing algorithm in Sowiport (terms) is the worst performing in Jabref. The best performing algorithm in Jabref (keyphrases) is 70 (click-through rate; 0.1

READ FULL TEXT
research
05/30/2018

One-at-a-time: A Meta-Learning Recommender-System for Recommendation-Algorithm Selection on Micro Level

In this proposal we present the idea of a "macro recommender system", an...
research
12/18/2019

Meta-Learned Per-Instance Algorithm Selection in Scholarly Recommender Systems

The effectiveness of recommender system algorithms varies in different r...
research
07/13/2017

Classifying document types to enhance search and recommendations in digital libraries

In this paper, we address the problem of classifying documents available...
research
11/26/2018

The Architecture of Mr. DLib's Scientific Recommender-System API

Recommender systems in academia are not widely available. This may be in...
research
06/17/2020

Quick Lists: Enriched Playlist Embeddings for Future Playlist Recommendation

Recommending playlists to users in the context of a digital music servic...
research
03/27/2017

Mr. DLib: Recommendations-as-a-Service (RaaS) for Academia

Only few digital libraries and reference managers offer recommender syst...
research
08/19/2020

Auto-Surprise: An Automated Recommender-System (AutoRecSys) Library with Tree of Parzens Estimator (TPE) Optimization

We introduce Auto-Surprise, an Automated Recommender System library. Aut...

Please sign up or login with your details

Forgot password? Click here to reset