no code implementations • WOSP 2020 • Divyanshu Marwah, Joeran Beel
Term weighting is responsible for computing the relevance scores and consequently differentiating between the terms in a document.
no code implementations • WOSP 2020 • Paul Molloy, Joeran Beel, Akiko Aizawa
The prediction can be used in the same way as real citation proximity to calculate document relatedness, even for uncited documents.
1 code implementation • 6 Feb 2024 • Tobias Vente, Joeran Beel
We found that AutoML and AutoRecSys libraries performed best.
no code implementations • 17 Jul 2023 • Lennart Purucker, Lennart Schneider, Marie Anastacio, Joeran Beel, Bernd Bischl, Holger Hoos
Automated machine learning (AutoML) systems commonly ensemble models post hoc to improve predictive performance, typically via greedy ensemble selection (GES).
no code implementations • 1 Jul 2023 • Lennart Purucker, Joeran Beel
Consequently, we compared the performance of covariance matrix adaptation evolution strategy (CMA-ES), state-of-the-art gradient-free numerical optimization, to GES on the 71 classification datasets from the AutoML benchmark for AutoGluon.
1 code implementation • 1 Jul 2023 • Lennart Purucker, Joeran Beel
Moreover, we present an example of using Assembled-OpenML to compare a set of ensemble techniques.
no code implementations • 30 Dec 2020 • Andrew Collins, Laura Tierney, Joeran Beel
To the best of our knowledge, this is the first effective meta-learning technique for per-instance algorithm selection in recommender systems.
1 code implementation • 10 Sep 2020 • Oisín Carroll, Joeran Beel
FGNNs are shown to improve the performance of networks playing checkers (draughts), and can be easily adapted to other games and learning problems.
no code implementations • 7 Sep 2020 • Mohammed Al-Rawi, Joeran Beel
As a result, AI companies are relying on manually annotated fashion data to build different applications.
1 code implementation • 19 Aug 2020 • Rohan Anand, Joeran Beel
Auto-Surprise is an extension of the Surprise recommender system library and eases the algorithm selection and configuration process.
no code implementations • 22 Jun 2020 • Joeran Beel, Bryan Tyrell, Edward Bergman, Andrew Collins, Shahad Nagoor
Our work includes a novel performance metric and method for selecting training samples.
no code implementations • WOSP 2020 • Mark Grennan, Joeran Beel
We find that both synthetic and organic reference strings are equally suited for training Grobid (F1 = 0. 74).
no code implementations • 18 Dec 2019 • Andrew Collins, Joeran Beel
User engagement was significantly increased for recommendations generated using our meta-learning approach when compared to a random selection of algorithm (Click-through rate (CTR); 0. 51% vs. 0. 44%, Chi-Squared test; p < 0. 1), however our approach did not produce a higher CTR than the best algorithm alone (CTR; MoreLikeThis (Title): 0. 58%).
no code implementations • 16 Dec 2019 • Nicholas Bonello, Joeran Beel, Seamus Lawless, Jeremy Debattista
Fantasy Premier League (FPL) performance predictors tend to base their algorithms purely on historical statistical data.
no code implementations • 16 Dec 2019 • Conor O'Sullivan, Joeran Beel
The heuristic achieved an overall test accuracy of 86. 68% which is 29. 7% higher than the models.
no code implementations • 15 Dec 2019 • Dominika Tkaczyk, Andrew Collins, Joeran Beel
In this paper, we present 1) A statistical analysis of roles in author contributions sections, and 2) Na\"iveRole, a novel approach to extract structured authors' roles from author contribution sections.
1 code implementation • WS 2019 • Mark Collier, Joeran Beel
Memory-augmented neural networks (MANNs) have been shown to outperform other recurrent neural network architectures on a series of artificial sequence learning tasks, yet they have had limited application to real-world tasks.
no code implementations • 27 May 2019 • Andrew Collins, Joeran Beel
There is a ~400% difference in effectiveness between the best and worst algorithm in both scenarios separately.
no code implementations • 26 Nov 2018 • Dominika Tkaczyk, Rohit Gupta, Riccardo Cinti, Joeran Beel
We propose ParsRec, a meta-learning based recommender-system that recommends the potentially most effective parser for a given reference string.
no code implementations • 26 Nov 2018 • Joeran Beel, Andrew Collins, Akiko Aizawa
In this paper, we introduce Mr. DLib's "Recommendations as-a-Service" (RaaS) API that allows operators of academic products to easily integrate a scientific recommender system into their products.
1 code implementation • 27 Sep 2018 • Mark Collier, Joeran Beel
Our experimental results provide an empirical basis for the choice of syllabus on a new problem that could benefit from curriculum learning.
6 code implementations • 23 Jul 2018 • Mark Collier, Joeran Beel
Our implementation learns to solve three sequential learning tasks from the original NTM paper.
no code implementations • 19 Jul 2018 • Joeran Beel, Andrew Collins, Oliver Kopp, Linus W. Dietz, Petr Knoth
We present the architecture of Mr. DLib's living lab as well as usage statistics on the first sixteen months of operating it.
no code implementations • 18 Jul 2018 • Joeran Beel, Barry Smyth, Andrew Collins
The main contribution of this paper is to introduce and describe a new recommender-systems dataset (RARD II).
no code implementations • 19 Feb 2018 • Andrew Collins, Dominika Tkaczyk, Akiko Aizawa, Joeran Beel
We conduct a study in a real-world recommender system that delivered ten million related-article recommendations to the users of the digital library Sowiport, and the reference manager JabRef.
no code implementations • 4 Feb 2018 • Dominika Tkaczyk, Andrew Collins, Joeran Beel
In this paper, we present an analysis of roles commonly appearing in the content of these sections, and propose an algorithm for automatic extraction of authors' roles from natural language text in scientific publications.
Digital Libraries